138 ANALYSIS OF MILK. 



basin ; when this occuts — after about an hour's drying — the 

 basin is so inclined that the fat runs down to the side away 

 from the solids not fat. Under these conditions drying is 

 completed in about five hours. It is an advantage to add an 

 equal volume of alcohol to the cream, and well mix before 

 drying, as the time of drying is thereby shortened. 



After weighing the total solids, the basin is replaced in the 

 water-oven for a few minutes to melt the fat ; 25 c.c. of amyl 

 alcohol are poured on, the basin again placed in the water-oven 

 for ten minutes, and the amyl alcohol solution of fat carefully 

 decanted while still hot ; with care, none of the solids not fat 

 passes away with the amyl alcohol. This process is repeated 

 eight times more, the basin being allowed to stand all night 

 between the fourth and fifth treatments. After the 'last treat- 

 ment, the amyl alcohol is drained off as far as possible, and the 

 basin and its contents dried for three hours in the water-oven. 

 The residue is weighed as solids not fat. This is now burnt 

 over a low flame, and the residue weighed as ash. Ether or 

 chloroform may be substituted for amyl alcohol, but the latter is 

 cheaper and less volatile. 



This method is not available for homogenised cream, and the 

 fat should be estimated by the Gottlieb or Werner-Schmid 

 methods. 



An indirect estimation of the percentage of fat may be made 

 from the total solids, and vice versd. 



For this purpose it is assumed that the proportion of solids 

 not fat to water in milk is constant, an assumption which causes 

 no appreciable error in cream analysis. It is found that on 

 the average 100 parts of water contain 10'2 parts of solids not 

 fat (see p. 150) ; we may assume that the water in cream con- 

 tains the same proportion of solids not fat, and estimate the 

 fat by deducting the percentage of water multiplied by 0"102 from 

 the total solids. 



The following formula will express this relation : — 



Let T = total solids and F = fat, 



then F = 1-102 T - 10-2. 



The following table (XIII.) may be used. 



For the determination of milk-sugar by polarisation, the cream 

 should be diluted with water ; 50 grammes may be made up 

 to 100 c.c, and 1 c.c. of acid mercuric nitrate added. 



If total nitrogen is determined, the cream should be evapor- 

 ated in a wide-mouthed flask, and the bulk of the fat extracted, 

 as, otherwise, great charring of the fat takes place (if the cream 

 be treated by the Kjeldahl method), and much carbon, difiicult to 

 dissolve, is produced. 



