CH. V.] METHOD OF AMOS PETERS. 139 



more than 180 mg. of glucose). Fit the two-holed rubber stopper 

 firmly into the neck of the flask, adjust the thermometer so that 

 its lower end is 2 mm. from the bottom of the flask and place on 

 the heated gauze. Note the time when the mercury indicates a 

 temperature of 95 C. Allow the heating to continue for exactly 

 20 sees, beyond this. Remove the flask by gripping the rubber 

 stopper and swill it for a second or two under the tap or in a bowl of 

 water. The reduction of the temperature practically stops the 

 reduction. Filter the hot fluid at once, using the stem of the ther- 

 mometer as a stirring rod. Wash the flask twice with about 7 cc. 

 of distilled water. Cool the filtrate by holding the flask under the 

 tap. Add exactly 4 cc. of strong sulphuric acid, insert a ther- 

 mometer and cool to 20. Add 6-5 to 7 cc. of the saturated solution 

 of potassium iodide, washing the stem of the thermometer with this 

 solution. Titrate at once with the standardised solution of sodium 

 thiosulphate as described above, using soluble starch as an indicator 

 when near the end point. 



Calculation of results. From the amount of thiosulphate 

 required the amount of copper in the filtrate is determined. Know- 

 ing the amount taken (352-9 mg.), the amount reduced by the sugar 

 can be calculated. The amount of glucose corresponding to this 

 copper can be determined by a reference to the curve in Fig. j 7. 



Example. The copper in the filtrate required 14-62 cc. of 

 thiosulphate. 



i cc. of thiosulphate = 12-86 mg. Cu. 



So copper in filtrate = 14-62 x 12-86 = 188-0 mg. Cu. 



So copper reduced by glucose in 20 cc.= 352 -9 188-0=164-9 mg. 



From the curve this is seen to correspond to 86-3 mg. glucose. 



So 20 cc. contain 86-3 mg. glucose. 



So 100 cc. contain 431-5 mg. glucose. = 0-431 per cent. 



NOTE. If the amount of reduced copper is between 60 and 200 mg., the 

 amount of glucose corresponding to this can be obtained by multiplying by 

 0-522. 



i6iA. The estimation of lactose by the copper-iodide method. 



The method is exactly similar to that described in the 

 previous exercise. The author is responsible for the 



