186 THE URINE. 



very small quantities of sugar can be detected. Many medicinal 

 substances pass into the urine and react with this test also. 



The phenyl-hydrazin test is not affected by the reducing 

 matters of the urine, but it gives a similar precipitate with milk- 

 sugar. Pure phenyl-hydrazin must be used. If it is the hydro- 

 chlorid, the crystals should be white, not brown. 



The fermentation test is not very sensitive. It may be inter- 

 fered with by the presence of some drugs which stop the action 

 of the yeast. If the urine is not acid, it should be made faintly 

 so with tartaric acid. It can be used to distinguish between glu- 

 cose and lactose. Barfoed's test (Experiment 28) may be employed 

 for the same purpose. 



389. Determine the quantity of sugar in diabetic 

 urine,, using Fehling's solution (Experiment 33). Dilute 

 the urine with a measured volume of water if necessary, 

 and use in the burette^ as was done in the case of the pure 

 glucose solution. 



ACETONE, (CH 3 ) 2 CO. 



Normally acetone is present in the urine only in 

 traces. Pathologically it occurs there in severe diabetes, 

 in fevers, in inanition, and cachectic conditions, as well 

 as in psychoses. In diabetes it often is a precursor of the 

 more dangerous diacetic acid. It appears to be formed by 

 the decomposition of albuminous compounds, and it can 

 be produced in the urine by the use of a diet of such 

 substances. It is a colorless liquid of a fruity odor, which 

 boils at 56.5 C. and which can consequently be readily 

 distilled from the urine. The examination of the urine 

 should be made while it is fresh. 



If a large quantity of acetone is present in urine the 

 latter may be tested directly. For small amounts it is best 

 to distill from about a liter one-fourth of its volume after 

 slightly acidifying with sulphuric acid. Place the distillate 



