276 PBACTICAL PHYSIOLOGY 



titration has to be carried out in a flask through the cork of which pass 

 two tubes, one from the burette, and the other to allow the ammonia 

 and steam to escape. 



Both these disadvantages are obviated by employing, instead of 

 ammonia, a solution of potassium cupro-cyanide. This has the power of 

 dissolving cuprous oxide, to form a colourless solution, and the reduced 

 oxide absorbs oxygen so slowly from the air that the titration may be 

 carried out in a basin. 



The solution of potassium cupro-cyanide is prepared by adding to 

 10 c.c. of boiling Fehling's solution, diluted with 40 c.c. of water, a 

 5 per cent, solution of potassium cyanide until the blue colour just 

 disappears. What happens is that the double cyanide of potassium and 

 copper is formed, which is colourless. Now, this salt is not reduced by 

 dextrose, but if it be mixed with more Fehling's solution it in no way 

 interferes with the reducing power of this, and at the same time it 

 dissolves the cuprous salt whenever it is formed. To carry out the 

 titration the above solution of potassium cupro-cyanide is mixed with 

 10 c.c. Fehling's solution, the further technique being exactly the same 

 as described in Experiment VIIL, the end reaction in this case being the 

 moment at which the solution becomes colourless. 



The solution, with the excess of Fehling's solution added, keeps very 

 well, and for clinical purposes, at least, should be prepared in stock. 1 



For very accurate determinations of sugar, the Allihn-Soxhlet 

 method should be used. The titration methods described above are 

 not quite accurate, because an excess of copper in a mixture of 

 Fehling's solution and glucose increases the reducing power of the 

 sugar. When a sugar solution is gradually run into a fixed quantity 

 of Fehling's solution, therefore, the reducing power of the sugar 

 becomes less and less as the titration proceeds, and the final result is 

 apt to be too low. This error is obviated in the Allihn-Soxhlet 

 method where a small quantity, 25 c.cm. of the sugar solution, which 

 must be of a percentage less than one, is discharged at once into a large 

 volume of diluted Fehling's solution (60 c.cm. + 60 c cm. water), and 

 the mixture boiled in an evaporating dish for exactly two minutes. 

 The cuprous oxide thus produced is quickly collected on the asbestos 

 weighing filter (p. 492) connected with a suction filter flask, the basin 

 thoroughly washed on to the filter with boiling water, the precipitate 

 washed with alcohol and ether, and then dried. 



The cuprous oxide is now reduced to metallic copper by passing a 



J The potassium cupro-cyanide may also be prepared by adding pot. cyanide 

 solution to a solution of cupric sulphate. For details of preparation of stock 

 solution see Button's Volumetric Analysis (ed. 1896), p. 317. 



