ESTIMATION OF JIILK-SUGAE. 97 



minutes. Filter tkrough a small filter, or, preferably, through. 

 a Gooch crucible, leaving the cuprous oxide as much as possible 

 in the beaker ; immediately the last drops of solution have been 

 poured on the filter, pour boiling well-hoiled water on the preci- 

 pitate, and wash several times by decantation. The precipitate 

 may also be separated and washed by centrifuging. Finally, 

 transfer the precipitate to the filter, and wash well with boiling 

 water. If a filter is used, transfer it to a small crucible, and 

 ignite over a very small flame till the filter is thoroughly charred, 

 and then gradually increase the flame till the highest available 

 temperature is obtained. It is better to use a porcelain crucible 

 than a platinum one, because platinum is permeable to reducing 

 gases from the flame, and complete oxidation cannot be obtained. 

 If a Gooch crucible is used it should be ignited in a muffle, and 

 the cuprous oxide thus converted into cupric oxide. The weight 

 of the cuprous oxide multiplied by 0'6024 will give the weight of 

 hydrated milk-sugar. If a filter is used, the weight of the ash 

 must, of course, bo deducted ; this should be obtained by ii,'iiiting 

 filters of the same kind, which have been treated in exactly the 

 same manner as in the estimation of milk-sugar, using the same 

 quantity of Fehlini^'s solution, but omitting the milk-sugar. 

 This is necessary, because the filter takes up mineral matter 

 from the Fehling's solution. If this precaution is taken, filtra- 

 tion through paper is satisfactory. 



The author finds it far more satisfactory, instead of employing 

 an arbitrary factor, which is not of absolute exactitude, to weigh 

 out a quantity of pure milk-sugar, approximating as nearly as 

 possible to that contained in the solution to be tested, and to 

 estimate the copper oxide obtained by treating it side b}' side 

 with the actual estimation. The extra trouble is nominal, and 

 the slight variations in the factor, due to dilution, etc., are 

 thereby compensated. 



(2) Wcln's Method. — The same quantities of solutions are used, 

 but the Fehling's solution, instead of being diluted with water, 

 is mixed directly with the milk-sugar solution, placed over a 

 naked flame, and raised as rapidly as possible to boiling ; boihng 

 is continued for exactly six minutes. The solution is filtered 

 through a tube about 1 cm. wide, constricted at the end, and 

 plugged with asbestos (Fig. 7). This tube is weighed, after 

 having been gently ignited, and the filtration is hastened by 

 means of a water pump ; a small funnel is used to pour the 

 solution into the tube. The precipitate of cuprous oxide is 

 washed with boiUng irell-hoUcd water, and transferred to the tube. 

 When the precipitate is well washed, the tube is sucked as dry 

 as possible by the pump. It is then detached from the filter 

 pump, and connected with a hydrogen apparatus, being clamped 



7 



