C. S. RAM AYYAR AND ZAL R. KOTHAVALLA                 135

                                      TABLE V

        Effect of varying the quantity of iso-butyl alcohol

From these results, it can be seen that the
amount of iso-butyl alcohol used, appeared to
have some effect on the reading. Amounts less
than 1 ml. or more than 1.2 ml. gave either low
or high percentage of fat and no or incomplete
separation. On the other hand, 1 ml. seems to be
the optimum amount which would give more or
less a correct reading.

(6) Effect of excess of alcohol on the reading of
the fat column. The slightly higher reading, which
was obtained when more than 1 ml. of alcohol was
used, according to Table V above, was suspected
to be due to some of the alcohol itself getting
separated along with the fat. To find out whether
that was actually the case, the following further
trials were conducted. Varying amounts of iso-
butyl alcohol, from 0.5 ml. to 1.5 ml. were added
to 5.5 ml. of salicylate reagent, followed by 10 ml.
of water. The tubes were well shaken and kept
at 60°C. The results obtained are shown in
Table VI.

                          TABLE VI

          Effect of using excess of alcohol

From Table VI, it is clear that the maximum
amount of alcohol miscible with 5.5 ml. of the
alkaline solution was 1 ml. Any amount in excess
of it separates out. This confirmed the previous
hypothesis that the higher readings with more
than 1 ml. of iso-butyl alcohol were due to the
separation of a portion of the alcohol itself.

(7) Effect of varying the quantities of salicylate
reagent used. For this purpose salicylate reagent
varying in quantities from 5.0—6.5 ml. was
mixed with a fixed amount of iso-butyl alcohol,
viz. 1 ml. The results obtained are given in
Table VII.

                        TABLE VII

Effect of using varying quantities of salicylate
                          reagent

The results show that slight excess over 5.5 ml.
of salicylate reagent did not materially affect the
readings, but if the quantity was less than this,
the reading obtained was lower.

(8) The Ram-Ko method for the determination of
fat in milk using an alkaline solution. The alkaline
reagent used in this method was prepared as
follows :—

Iso-butyl alcohol was not incorporated in this
solution, as it was observed that the alcohol
separated out into a layer at the top and shaking
the solution before use did not bring about a
thorough mixture to ensure the accurate amount
of alcohol essential for the test. It has already
been indicated how necessary it is that the amount
of alcohol added should be accurately measured
and it should be exactly 1 ml., as slight variations
lead to incorrect readings. Hence the two solu-
tions were kept separate. This also ensured a
long life for the solutions.

Adopting the above, the fat estimation in milk
can be carried out accurately as follows :—

First, 5.5 ml. of salicylate reagent are placed in
the butyrometer (Fig. 1). To this is then added
accurately by means of a pipette 1 ml. of iso-
butyl alcohol and 9.7 ml. of milk. The butyro-
meter is then corked and the contents well shaken
and thoroughly mixed. The tube is then kept in a
water bath at 60°C. with the cork end resting on