SUGAR. 761 



may lead to negative or inaccurate results. In this case proceed in the 

 following way: Treat the acid urine, or urine which has been faintly 

 acidified with a little sulphuric acid, with yeast whose activity has been 

 tested by a special test on a sugar solution, and allow it to stand 24-30 

 hours at about 30. Then test again with the bismuth test, and if the 

 reaction now gives negative results, then sugar was previously present. 

 But if the reaction continues to give positive results, then it shows, 

 if the yeast is active, the presence of other reducing, unfermentable 

 substances. 



In performing the fermentation test care should be taken that the urine 

 be acid before as well as after fermentation. If the reaction becomes 

 alkaline during fermentation (alkaline fermentation), then the test 

 must be discarded. The vessel must be perfectly clean and strongly 

 heated before use. To make sure the urine may be boiled before fer- 

 mentation. 1 



If a good polariscope is at hand it must not be forgotten to control 

 the results of the fermentation by determining the rotation before and 

 after fermentation. The phenylhydrazine test also, in many otherwise 

 doubtful cases, gives good service in testing urines for sugar. 



Phenylhydrazine Test. According to v. JAKSCH this test is performed in 

 the following way : Add in a test-tube containing 6-8 cc. of the urine two 

 knife-pointfuls of phenylhydrazine hydrochloride and three knife-point fuls 

 of sodium acetate, and when the salts do not dissolve on warming add 

 more water. The test-tube is placed in boiling water and warmed on 

 the water-bath. It is then placed in a beaker of cold water. If the 

 quantity of sugar present is not too small, a yellow crystalline precipitate 

 is now obtained. If the precipitate appears amorphous, there are found, 

 on looking at it under the miscroscope, yellow needles singly and in groups. 

 If very little sugar is present, pour the test into a conical glass and examine 

 the sediment. In this case at least a few phenylglucosazone crystals 

 are found, while the occurrence of larger and smaller yellow plates or 

 highly refractive brown globules does not show the presence of sugar. 

 This reaction is very reliable, and by it the presence of 0.03 per cent 

 sugar can be detected (ROSENFELD, GEYER 2 ) . In doubtful cases where 

 certainty is desired, prepare the crystals from a large quantity of urine, 

 dissolve them on the filter by pouring over them hot alcohol, treat the 

 filtrate with water, and boil off the alcohol. Still better, the precipitate 

 is dissolved, according to NEUBERG, in some pyridine, and again precipi- 

 tated as crystals by the addition of benzene, ligroin, or ether. If the 

 characteristic yellow crystalline needles, whose melting-point (204- 

 205 C.) may also be determined, are now obtained, then this test is 

 decisive for the presence of sugar. It must not be forgotten that levulose 

 gives the same osazone as dextrose, and that a further investigation is 

 necessary in certain cases. 



The following modification by A. NEUMANN is simple, practical, and at the 

 same time sufficiently delicate. 5. cc. of the urine are treated with 2 cc. of acetic 



1 On the performance of the fermentation test and certain sources of error, see 

 Salkowski, Berlin, klin. Wochenschr., 1905 (Ewald-Festnummer), and Pfliiger, Pfliiger's 

 Arch., 105 and 111. 



2 Rosenfeld, Deutsch. med. Wochenschr., 1888; Geyer, cited from Roos, Zeitschr. 

 f. physiol. Chem., 15. . 



