252       Carotene and Vitamin-A Contents of butter-fat in cows

of the fat. As Xanthophyll only constitutes 1/15th of the total yellow colour
of butter-fats, no attempt was made for its estimation [Ferguson and Bishop,
1936].

Determination of Vitamin-A content of butter.—By carrying out some
preliminary experiments, it has been found that the blue-value on the un-
saponifiable fraction gives higher and more concordant results than the value
without saponification, probably because of the removal of colour inhibitors.
After a number of trials the following method has been found to give satisfac-
tory results.

Procedure.—About 10 grm. of the clear and filtered fat was weighed into
an Erlenmeyer flask and then saponified with an equal amount of 66 per cent
alcoholic KOH solution and 20 c.c. of alcohol. The process was carried out
electrically for half an hour carefully. The flask was then cooled and 75 c.c.
of the distilled water added. The contents were poured into a separating fun-
nel and the flask washed with 100 c.c. of ether (peroxide free) and then poured
into the funnel. On carefully shaking the funnel the fatty residue, carotene
and vitamin-A present in the digest passed into the ether layer. Sometimes
it happened that the ether layer did not separate well and the whole mass
formed an emulsion. This was prevented by adding a little alcohol when
the ether layer separated out quickly. This was then separated, washed
twice with distilled water, 3 per cent KOH solution, 1 : 10 HC1 and finally
two times with water and it was then dried by filtering through a layer of
anhydrous sodium sulphate (B. D. H.) and evaporated, in an atmosphere of
carbon dioxide. The resultant fatty residue was at once dissolved in 1 c.c. of
chloroform. To 0. 2 c.c. of the solution was added 2 c.c. of SbCl₃ solution (30
per cent solution of SbCl₃ in dry CHCl₃) and the blue colour matched quickly
in a Lovibond tintometer. Results were then calculated in Moore's units
as follows :—

Example.—10 grm. (nearest to 0.1 grm.) of butter were digested and the
extracted residue was dissolved in 1 c.c. of CHCL₃. 0.2 c.c. of this solution
was treated with 2 c.c. of SbCl₃ reagent, which gave a reading of 3.8 B. U.,
0.6 Y. U. and 0.7 R. U. Then the blue unit per grm. of the butter-fat is given
by 3.8 × 2.2 / 0.2 × 10 = 4.10 Lovibond Blue Units (B. U.) per grm. of the butter-fat
corresponding to a 1 cm. thickness of the fat.

The estimation of carotene in milk, grass and fodders was carried out by
the method of Ferguson & Bishop [1936].

All the vitamin-A contents of butter-fats set out in the tables are given in
Lovibond Blue Units (B. U.) and carotene contents in Lovibond Yellow Units
(Y. U.) per grm. of the fat corresponding to a 1 cm. thickness of the fat. The
determinations of carotene and vitamin-A in butter were always carried out
in duplicate. The results given in following tables represent the mean of 2
readings which did not differ by more than 0.2 Y. U. or B. U.