View report page

118
Annual Report of the London County Council, 1910
In eleven cases portions of the milk samples were kept in the cold room, until after the death
of the guinea-pigs injected with the deposits from the samples and the identification of the bacilli
derived from the dead animals. From the milk so kept, in four instances, there was recovered the
same organism as that found in the guinea-pigs' heart or spleen or in the inflamed tissue at the site
of inoculation. The bacilli recovered in these instances were B. coli communis, B. griinthal, B. Friedlander,
and No. 7 in MacConkey's list.
Bacilli of the Proteus group were found occasionally but were probably post-mortem invaders
of the organs from the intestine, since they are non-pathogenic to guinea-pigs.
The cocci found were also not pathogenic and cannot be regarded as having any causal connection
with the deaths.
Consideration of the number of organisms injected with the milk deposits.—Having obtained some
idea of the kind of organisms present in the deposits an estimate of the number would be of interest.
Organisms lethal to guinea-pigs are present in the deposits, but it is not easy to ascertain if sufficient
of them to constitute a fatal dose are present in any given deposit, by any other method than that of
injecting it into the guinea-pigs—a method which is open to many fallacies (see Conclusions, p. 121).
But even if it were possible to estimate the total number of each variety of pathogenic organism present
in a given deposit, it would still be extremely difficult to ascertain the total effect of different doses
of different bacilli, especially when we consider that the deposit contains in addition a large mass of
foreign proteid matter in the form of cells, milk proteids and non-pathogenic organisms. Practically,
the problem is an insoluble one on account of the technical difficulties.
Pathogenicity tests.—In testing the pathogenicity of the organisms isolated, the question of dose
is an important one. A number of pathogenicity tests were done with various bacilli isolated. The
proteus organisms and cocci are not pathogenic, but all the lactose-fermenting organisms may be
expected to kill a guinea-pig, if injected subcutaneously, in doses of 2 or 3 c.c. of a broth culture kept
for two days at 37 deg. C. But, since we cannot actually count the numbers of each kind of bacillus
present in any milk deposit, we cannot decide whether a lethal dose of a given organism, or a lethal
dose of each of several organisms, was present in the deposit injected. One can at least be fairly certain
that in some samples fatal doses of bacilli were injected, since the same bacillus was found in the milk
and in the guinea-pig and belonged to the group of lactose fermenting organisms fatal to guinea-pigs
in doses of 1 or 2 c.c. of a 24-hour broth culture.
Experiments to determine the content of "lactose fermenting organisms" (indicating fæcal pollution)
in milk samples and its relation to the lethal effect of the injected sediment.—The number of lactose
fermenters in the various milk samples can be roughly deduced by the fermentation test. Small
quantities (.001 c.c. or .01 c.c.) of the various samples of milk were inoculated into glucose-bile-saltbroth,
and those tubes in which fermentation, i.e., gas-formation, occurred were noted. This gas
formation is roughly an evidence of the presence of lactose fermenting organisms in sufficient numbers
in the amount stated to produce gas in 24 hours.
In the course of these tests it seemed advisable to see if the absence of fermentation in the tubes
showing no gas was due to the presence of a preservative in the milk.
Tests for the presence of preservatives were carried out in the case of 54 samples by doing two
parallel experiments. In one series, (a) a quantity of .01 c.c. of each milk sample was inoculated into
glucose bile salt broth, and the whole set of tubes incubated at 37 deg. C. In the other, (6), 5 c.c. of
each sample were inoculated into 3 c.c. of glucose bile salt broth, and to each tube a few drops of a
dilute broth culture of an organism known to ferment glucose were added. If any considerable
quantity of a preservative had been present in a milk sample, it would probably have killed, or at least
have prevented the growth of, the added organisms, so that in such a case any tube that did not
ferment in series (a), and that failed to ferment in series (6), would raise a strong suggestion of the
presence of preservative in the corresponding milk sample.
It was satisfactory to find that in not one of the 54 samples tested by this method was there
any evidence of the presence of preservatives in amount sufficient to inhibit bacterial growth, so that
the fermentation results recorded (see Diagram) can be accepted as a rough measure of the numbers of
lactose fermenters present.
With regard to the fate of the animals inoculated respectively with deposits from the "nonfermenting"
and from the "fermenting" milks, the numbers are not very large, it is true, but it is
somewhat striking that the proportion of deaths (34 in 322 and 6 in 60) is in each case practically the
same—whereas one would have been inclined to expect the guinea-pigs, corresponding to the " gas "
or "fermentation milks, to show a larger death-rate than those inoculated from the "non-fermentation"
milks. This is one of the many illustrations evident in this enquiry of the extrmee complexity
of the various factors involved.
Results of subcutaneous inoculation of sour milk.—(i.) Untreated; (ii.) after filtration; (iii.) after
being boiled. The injection of a few cubic centimetres of milk which has become sour is usually fatal
to guinea-pigs. The toxicity of the sour milk might conceivably be due (a) to the pathogenicity of
some of the bacteria contained therein, or (b) to the poisonous qualities of the not inconsiderable amount
of bacterial proteins contained in the bacilli injected, supposing the bacteria to be themselves incapable
of surviving in the animals' tissues, and (c) to the presence of soluble toxins from the bacteria. Filtering
the milk would remove all bacteria, and therefore factors number 1 and 2. Boiling the milk would
kill the bacteria, but at the same time diminish or destroy the poisonous properties of the contained
organisms and probably also any soluble toxin. The experiments eliminate the third factor above,
namely, presence of soluble toxin as a cause of the mortality, but do not determine the relative
importance of the other two factors in contributing to the death of the animals.