i MILK 85 



precipitate washed with saturated magnesium sulphate solution, 

 and the nitrogen estimated in it by Kjeldahl's method. To 

 obtain the amount of casein corresponding to the nitrogen 

 found, the factor 6*38 is employed. 



3. Estimation of Albumin. 



The albumin in milk can be estimated by simply taking the 

 difference between the amounts of protein and casein, or it is 

 easy to make a direct determination by taking the filtrate got 

 from the precipitation of the casein in Sebelien's method (see 

 above), diluting with water, and precipitating the albumin 

 by means of tannic acid. The nitrogen can then be deter- 

 mined in the precipitate by Kjeldahl's process, and the result 



multiplied by 6'34 = 



V. ESTIMATION OF THE MILK SUGAR. 

 1. Gravimetric Methods. 



The gravimetric methods for the estimation of milk sugar 

 are based upon the power which it possesses of reducing an 

 alkaline copper solution (Fehling's solution), whereby an 

 amount of cuprous oxide equivalent to the weight of milk 

 sugar is formed. With milk sugar, however, the relation 

 between the two is not always proportional ; a small amount 

 of sugar gives a quantity of cuprous oxide which is relatively 

 smaller than when large amounts of milk sugar are used 

 Soxhlet has therefore constructed a table (Table V in Appendix) 

 in which these discrepancies are corrected. The cuprous oxide 

 is not weighed as such, but converted either into copper or 

 cupric oxide. 



(a) Allihn- Soxhlet Method. 



Twenty-five c.c. of milk are weighed out, and treated in exactly 

 the same manner as in Ritthausen's method for estimating 

 the total protein in milk (see p. 83). After the addition of 

 the alkali the liquid should be neutral or slightly acid ; it 

 should not under any circumstances have an alkaline reaction. 

 Scheibe's l . modification, which is to add 20 c c. of a cold 

 1 Milch-Zeitung, 1901, p. 113. 



