204 QUANTITATIVE ANALYSIS. 



centimeter of 1/10 normal thiosulphate solution corresponds 

 to 6.36 mg. of copper. 



If the quantity of cuprous oxide which separates is very 

 small it may be well washed, dissolved in nitric acid, the 

 nitrous acid removed by means of urea, and the copper 

 directly determined. The results may easily be too high. 

 (K. B. Lehmann.) 



Very frequently in the case of urines containing a small 

 amount of sugar the cuprous oxide does not precipitate. 

 In this case neither the titration method nor the gravimetric 

 method can be used. For such cases the following solu- 

 tion is recommended by Arthus: 125 cc. of Fehling's solu- 

 tion and 5 g. of ferrocyanide of potassium are diluted to one 

 liter. Eight cubic centimeters of this solution correspond 

 to 1 cc. of Fehling's solution. This solution is only decolor- 

 ized by the glucose ; however, the recognition of the end-point 

 is rendered very uncertain by the color of the mixture. 



Normal urine contains reducing substances correspond- 

 ing to about 0.2 to 0.3 per cent, of glucose. If, therefore, we 

 titrate a normal urine to which a known quantity of glucose 

 has been added, the determination will of course give results 

 correspondingly too high. For diabetic urine this error is 

 apparently not so great and need not be considered, espe- 

 cially when the urine is diluted. 



The determination of the sugar from the volume of the 

 carbon dioxide evolved on fermenting the urine is very con- 

 venient, though less accurate. For this purpose we may 

 use the empirically graduated "Fermentation Sacchari- 

 meter" of Einhorn, Fiebig, or Lohnstein, the latter also for 

 undiluted urine (Allg. med. Centralzeitung, 1899, No. 101, 

 and Munch, med. Wochenschr. 1899, No. 50). 



