REPORT OF THE KING INSTITUTE, GUINDY

37

the Analyst would probably have to be present in court, and a case which should be quite
simple might become complicated and expensive, so much so that after a little experience of
such cases some municipalities might refuse to go to the expense of bringing their Analyst
a long distance to give evidence in a case in which the vendor might not be in a position
to pay the costs in the event of a conviction. The simple certificate which. I was in the
habit of using in England, which merely stated that the sample under examination contained
a certain percentage of foreign fat (no analytical constants being given) leads to no such
complications.

   I may summarise my views on this point by saying that, when a very low Reichert value
is obtained (much below any standard which could possibly be fixed by law) this figure is in
itself quite sufficient to condemn a sample of butter-fat, but that when a value near the
border line is obtained an opinion can only be given by an Analyst with high qualifications
and with a thorough experience in the analysis of oils and fats (probably the most difficult
branch of analytical chemistry) and that he must be capable of carrying out accurately the
determination of a considerable number of constants, of making tests for specific adulterants
and of forming an opinion by the consideration. of the whole of his results. No analyst of
any standing would ever condemn. a sample unless he were fully convinced of the presence
of an adulterant and it goes without saying that the Act is only workable if the public
analysts appointed under it are men whose qualifications are beyond criticism.

   In Bengal standard Reichert values have been. fixed of 24 for cow butter-fat and 30
for buffalo. I do not know on what results these figures were based, probably the animals in
Bengal give very different figures to those in Madras. In Madras these limits would be
quite unworkable. Fifty-two samples of cow's milk butter-fat have been examined. Twenty-
five of these gave Reichert values below the Bengal limit of 24, four of these being
below 20, the lowest value being 14.7. Of 29 samples of buffalo-milk butter-fat only
two gave values which would enable them to be passed as genuine in Bengal, giving values
between 30 and 31, the remainder giving values . varying from 23.0 to 29.7. The average
value for cow's milk butter-fat was 23.1, for buffalo 26.7.

   After my draft regulations under clause 20 (e) I suggest draft regulations under 20 (d)
"prescribing the manner in which notice of any addition, admixture or deficiency shall be
given to the purchaser of any article of food."

   Here I am in some little difficulty. There are no such general regulations in England
which I can use as a basis; separate regulations having been prescribed in connexion with
different food-stuffs. Thus labelling regulations are laid down for margarine (butter
substitute) in the Butter and Margarine Acts. Other regulations prescribing methods of
giving notice of admixture occur in the Milk and Cream Regulations, Condensed Milk
Regulations, Dried Milk Regulations, these being regulations made under the Public Health.
Acts. As far as possible I have used these as a basis and my draft gives a general idea of
the type of notice which. I consider desirable, but it will be necessary for the advice of some
one fully conversant with Indian bazaars and retail shops to consider whether my suggested
regulations would be applicable to Indian conditions, advice which I am not competent to
give, and further it will be necessary to prescribe the languages in which the notices should
be printed in the various districts of the Presidency. I would advise that the regulations
should insist (as all the English regulations do) on all notices being printed.