View report page

93
milk for some considerable time, and there is no doubt
that the milk from this farm must have been contaminated
with a considerable quantity of pus and mastitis organisms.
The Medical Officer of Health for Monmouthshire
reported that examination of samples showed that on the
whole milk from that district was fairly satisfactory, but
that further investigation was being made.
I am of opinion that the facts as set out strongly confirm
the case, which I have put forward on previous
occasions, for the compulsory pasteurisation of all milk.
In no other way can a safe supply be assured.
BACTERIOLOGICAL EXAMINATION OF MILK.
In my Annual Report for 1934 I called attention to
the unsuitability of the plate count as a standard for grading
milk, and drew attention to the unreliability of reports
on the bacteriological examination of samples. I set out
in some detail the results of certain tests which I had had
carried out, and showed that reports from different laboratories
of the examination of samples of milk taken from
one bottle gave figures varying from 9.270 to 3,400,000
per c.c., and that the results obtained by one laboratory
varied from 147.000 to 3,400.000 for the same milk.
In October, 1935, Professor G. S. Wilson's valuable
report, "The Bacteriological Grading of Milk," was issued
by the Medical Research Council. In this book, which
sets out the results of three years of investigation, the
plate count is severely criticised. "Even under favourable
conditions, with the method standardised as far as
possible, the experimental error is very large, and on any
one count an allowance of ± 90 per cent. may have to
be made. Even this margin of error will not include all
results. . . . . It is therefore recommended that the plate
count should be discontinued as a method of grading
ordinary raw milk. . . . .The quantitative expression of
the results in figures extending over a wide range affords