ESTIMATION OF LECITHIXS. 137 



soda solution. This figure varies in normal milks from 18 -1" 

 to 22'6°, and averages 19'9° ; in fresh milks it is practically 

 equal to the acidity. 



The aldehyde figure obtained with strontia solution multiplied 

 by 0'170 will give a close approximation to the percentage of 

 proteins ; it is not an absolutely exact measure of the proteins, 

 as casein and albumin do not give the same aldehyde figure, 

 and the relative proportion of these is liable to slight variation^. 



The acidity and aldehyde figure usually approximate in fresh 

 milk to the same figure, it being only rarely that the difference 

 amounts to more than 2° ; in certain abnormal samples, usually 

 low in solids not fat, the acidity is much lower than the aldehyde 

 figure, and in these the factor 0'170 for calculating proteins is not 

 exact. 



Estimation of Lecithins. — Bordas and de Raczkowski deter- 

 mine lecithins in milk thus : — 100 o.c. of milk are shaken with 

 100 c.c. 95 per cent, alcohol, 100 c.c. of water and 10 drops of 

 acetic acid. The precipitate is extracted with three successive 

 quantities of 50 c.c. each of hot absolute alcohol. The mixed 

 extracts are evaporated, and the residue taken up with a small 

 quantity of a mixture of equal parts alcohol and ether ; the ether 

 is evaporated, and the residue saponified by potassium hydroxide, 

 and the soap solution acidified with dilute nitric acid. The fatty 

 acids are filtered off, and the filtrate evaporat(?d to dryness, 

 mixed with 10 c.c. of strong nitric acid, and completely oxidised 

 by the addition of potassium permanganate little by little. A 

 few drops of sodium nitrite solution (1 : 10) are added to dissolve 

 the manganese hydroxide, and the nitrous acid expelled by boiling. 

 The phosphoric acid is precipitated with ammonium molybdate 

 and estimated as magnesium pyrophosphate. The weight of 

 Mg.P.jOy multiplied by r5495 will give the glycero-phosphorie 

 acid, and by 7 '27 the lecithins. 



The Analysis of Milk Products. — For the analysis of milk 

 products, the methods described above can generally be used. 

 The following notes will show where it is advisable to depart 

 from them or employ modifications. 



Homogenised Milk. — The Adams or Ritthausen methods 

 cif fat estimation should not be used. 



Sterilised Milk. — The fat cannot be estimated by the 

 Ritthausen method. 



Cream. — The methods given above may be employed in the 

 analysis of cream. The following method of determining total 

 solids, fat (by difference), solids not fat, and ash is convenient. 



Four to five grammes are weighed in a wide platinum basin, 

 which is placed in a water-oven, till the water has apparently 

 evaporated and the solids not fat stick to the bottom of the 



