QUANTITATIVE ESTIMATION OF PROTEID IN URINE. 793 



in the spinal marrow. It gives a precipitate on heating to 40-60 C., 

 which on further heating to boiling dissolves again more or less completely, 

 depending upon the reaction and upon the amount of salt present. In 

 salt-free solution the precipitate is not dissolved, on heating to. boiling, 

 at least not always. It does not separate on dialysis, but can be pre- 

 cipitated from the urine by double the volume of a saturated ammonium- 

 sulphate solution or by alcohol. It has also been obtained as crystals 

 (GRUTTERINK and DE GRAAFF, MAGNUS-LEVY x ). This body shows a 

 varying behavior in the different cases in which it has been found and its 

 nature has not been explained. From the investigations of the above- 

 mentioned and other experimenters (MOITESSIER, ABDERHALDEN and 

 ROSTOSKI) we can draw the conclusion that this proteid is similar to the 

 proteoses in several reactions, but that nevertheless it stands close to the 

 genuine protein bodies. It also yields primary as well as secondary 

 proteoses on peptic digestion (GRUTTERINK and DE GRAAFF), and yields 

 the same hydrolytic cleavage products as the other proteins (ABDERHALDEN 

 and ROSTOSKI). 



Quantitative Estimation of Proteid in Urine. Of all the methods pro- 

 posed thus far, the COAGULATION METHOD (boiling with the addition of 

 acetic acid) when performed with sufficient care gives the best results. 

 The average error need never amount to more than 0.01 per cent, and it 

 is generallv smaller. With this method it is best to first find how much 

 acetic acid must be added to a small portion of the urine, which has been 

 previously heated on the water-bath, to completely separate the pro- 

 teid so that the filtrate will not respond to HELLER'S test. Then coagulate 

 20-50-100 cc. of the urine. Pour the urine into a beaker and heat on 

 the water-bath, add the required quantity of acetic acid slowly, stirring 

 constantly, and heat at the same time, Filter while warm, wash first 

 with water, then with alcohol and ether, dry and weigh, incinerate and 

 weigh again. In exact determinations the filtrate must not give HEL- 

 LER'S test. 



The separate estimation of GLOBULINS and ALBUMINS is done by carefully 

 neutralizing the urine and precipitating with MgSO 4 added to saturation (HAMMAR- 

 STEN), or simply by adding an equal volume of a saturated neutral solution of 

 ammonium sulphate (HOFMEISTER and POHL 2 ). The precipitate consisting of 

 globulin is thoroughly washed with a saturated magnesium-sulphate or half- 

 saturated ammonium-sulphate solution, dried continuously at 110 C., boiled 

 with water, extracted with alcohol and ether, then dried, weighed, incinerated, 

 and weighed again. The quantity of albumin is calculated as the difference 

 between the quantity of globulin and the total proteids. 



Approximate Estimation of Proteid in Urine. Of the methods suggested for 

 this purpose none has been more extensively employed than ESBACH'S. 



1 Magnus-Levy, Zeitschr. f. physiol. Chem., 30 (literature); Grutterink and de 

 Graaff, ibid., 34 and 36; Moitessier, Compt. rend. soc. biolog., 57; Ville and Derrien, 

 ibid., 62; Abderhalden and Rostoski, Zeitschr. f. physiol Chem., 46; see also Hopkins 

 and Savory, Journ. of Physiol., 42. 



2 Hammarsten, Pfliiger's Arch., 17; Hofmeister and Pohl, Arch. f. exp. Path. u. 

 Pharm., 20. 



