GRAVIMETRIC ESTIMATIONS 89 



heated to boiling; 25 c.c. of the sugar solution, so prepared as 

 not to contain more than 0^25 gram of glucose, are then 

 added, and the boiling is continued for exactly two minutes ; * 

 at the end of this time the supernatant liquid should be still 

 blue ; if not, the sugar solution was too strong and a fresh ex- 

 periment must be started using a more dilute sugar solution. 

 The liquid is then filtered through a weighed asbestos Gooch 

 crucible which has been previously washed first with water and 

 then successively with 10 c.c. of alcohol and a like quantity 

 of ether, and has been dried for half an hour in a steam oven 

 before weighing. The precipitate is then similarly washed 

 with hot water and finally with 10 c.c. of alcohol and 10 

 c.c. of ether, and is dried for half an hour in a steam oven. 

 After cooling in a desiccator, the Gooch crucible is weighed 

 again. 



The weight of cuprous oxide multiplied by the factor 

 0-8883 gives the weight of copper, from which the amount of 

 dextrose may be calculated by reference to the table on 

 pp. 87-88. 



An alternative method consists in mixing as before 30 c.c 

 of each of the solutions A and B, diluting them with 60 c.c. of 

 water and heating by immersing the mixture in a boiling 

 water bath for six minutes ; 25 c.c. of the sugar solution, con- 

 taining not more than 0*25 gram of glucose, are then boiled 

 and added to the copper solution ; the mixture is then heated in 

 the boiling water bath for another ten minutes. The precipi- 

 tated cuprous oxide is thereupon filtered, as before, on a tared 

 asbestos Gooch crucible, washed, dried, and weighed. 



As the results obtained by different workers are liable to 

 vary somewhat, it is best for each worker to determine for him- 

 self what results he obtains when using a glucose solution of 

 known strength ; by dividing the weight of glucose known to 

 be in the solution by the weight of cuprous oxide obtained, a 

 factor is found which on subsequent occasions can be used 

 for multiplying into the weight of cuprous oxide obtained, in 

 order to give the corresponding amount of glucose. 



A sufficiently accurate result can, however, usually be ob- 

 tained by employing the factors given in the following table for 



* According to Pfliiger this is not sufficiently long. 



