View report page

87
The presumptive standard requires a minimum of 8.5 per cent. of solidsnot-fat.
The system of classification adopted is as under:—
Solids-not-fat, 8.43 per cent. and over—Genuine.
Solids-not-fat, 8.25 per cent. to 8.42 per cent.—Poor quality.
Solids-not-fat, less than 8.25 per cent.—Adulterated.
During the year 26 samples have been classified as " Poor Quality." In
every case a letter has been sent by the Medical Officer of Health to the
vendor asking for an explanation of the deficiency of solids-not-fat.
One vendor produced a number of analytical certificates showing that the
samples were genuine. The figures given for percentage of solids-not-fat were
about 0.5 per cent. higher than those obtained in this laboratory.
As an example the laboratory figures obtained for one poor quality
sample were:—Fat 3.15 per cent. and solids-not-fat 8.38 per cent. The figures
produced by the vendor were:—Fat 3.20 per cent. and solids-not-fat 8.64 per
cent. As similar differences in analytical results were shown on another
occasion by the same vendor, a sample of milk was analysed and a portion
sent to the analyst in question. The laboratory figures were:—Fat 2.75 per
cent. and solids-not-fat 8.20 per cent. The analyst's figures were:—Fat 278
per cent. and solids-not-fat 8.64 per cent.
The vendor also showed results of analysis of genuine samples which had
been purchased from him formally, the percentage of solids-not-fat being given
in some cases as 9 4 and even higher. These results are obviously absurd.
Figures obtained by different analysts for percentage of solids-not-fat generally
agree to within 0 05 per cent. so that it will be seen that the large
discrepancy shown in the above results must have been caused by a serious
and constant error.
The vendor was advised to check the results by sending the samples to a
Public Analyst. It was then found that the results of analysis which he had
been previously obtaining were incorrect, and it was admitted later by the
analyst in question that his results were incorrect owing to the use of
inaccurate apparatus. These faulty analyses were not carried out by a
qualified analyst.