THE URINE. 397 



positive as HELLER'S test. The albumoses are also precipitated 

 by this reageut. 



Reaction with Acetic Acid and Potassium Ferrocyanide. Treat 

 the urine first with acetic acid of about 2$ and then add drop by 

 drop a potassium ferrocyanide solution (1 : 20), carefully avoiding 

 an excess. This test is very good, and in the hands of experts it 

 is even more delicate than HELLER'S. In the presence of very 

 small amounts of albumin it requires more practice and dexterity 

 than HELLER'S, as the relative quantities of reagent to the albumin 

 act on the result of the test. The amount of salts in the urine 

 also seems to have an influence. This reagent also precipitates 

 albumoses. 



The different color reactions cannot be directly used, especially 

 in deep-colored urines which only contain little albumin. The 

 common salt of the urine has a disturbing action on MILLON'S 

 reagent. To prove more positively the presence of albumin, the 

 precipitate obtained in the boiling test may be filtered, washed, 

 and then tested with MILLON'S reagent. The precipitate may also 

 be dissolved in dilute alkali and the biuret test applied to the solu- 

 tion. The presence of albumoses or peptones in the urine is 

 directly tested for by this last-mentioned test. In testing the 

 urine for albumin one must never be satisfied with one test alone, 

 but one must at least apply the heat test and HELLER'S test or the 

 potassium-ferrocyanide test. In using the heat test alone the 

 albumoses may be easily overlooked, but these are detected, on the 

 contrary, by HELLER'S test. If we are satisfied with this last test 

 or the potassium-ferrocyanide test alone, we have no sufficient inti- 

 mation of the kind of albumin present, whether it consists of albu- 

 moses or coagulable albumin. 



For practical purposes several dry reagents for albumin have been recom- 

 mended. Besides the metaphosphoric acid may be mentioned : PAVY'S re- 

 agent, which consists of small disks or plates of citric acid and sodium ferrocy- 

 anide ; STUTZ'S or FURBRINGER'S gelatine capsules, which contain mercuric 

 chloride, sodium chloride, and citric acid; and GEISSLEB'S albumin-test papers, 

 which consist of strips of filter-paper which have been dipped in a solution 

 of citric acid and also mercuric chloride and potassium-iodide solution and 

 then dried. 



If the presence of albumin has been positively proved in the 

 urine by the above tests, it then remains necessary to determine the 

 variety. 



The detection of globulin and albumin. In detecting serum- 

 globulin the urine is exactly neutralized, filtered, and treated with 

 magnesium sulphate in substance until it is completely saturated 

 at the ordinary temperature, or with an equal volume of a saturated 

 neutral solution of ammonium sulphate. In both cases a white, 

 flocculent precipitate is formed in the presence of globulin. In 



