View report page

76
the purification of sewage before proceeding to add special bacteria to a liquid or to a bacterial bed
already harbouring all the known microbes of putrefaction and nitrification.
Such contentions are certainly justifiable, but it is to be thought of that the path to knowledge
does not always lie along the most direct route, and accident has before now revealed the
truth where experiments conducted on apparently sound lines have entirely failed.
Although, as regards the so-called biological treatment of sewage, the results obtained in
the past have been most encouraging, at all events, from the chemical point of view, it is now
clearly established that all the different processes which have been tried, suffer from serious limitations
which at the present time seem almost insuperable in character.
It is conceivable that both from the practical and epidemiological point of view a solution
of the problem of sewage disposal may lie in one or other of the directions that has been indicated.
The stress of work involved in the periodical examination of the crude sewage and effluents
from the coke-beds made it impossible to carry out any prolonged research of the above nature.
Nevertheless, one experiment was tried, and although attended with apparently negative
results, its description will not be out of place.
It will be remembered that in previous Reports it was shown that a rapidly-liquefying gasforming
proteus-like germ was present in London crude sewage in numbers usually exceeding
100,000 per c.c. This microbe was also found to be present in the effluents in very large numbers.
Its ability to form gas, its rapid peptonising action and its abundance seemed to indicate that it
was possibly a microbe specially concerned in the work of purification. Accordingly it was
determined to isolate a strain of this micro-organism from an effluent, to cultivate it on a large
scale, and to add it in great amount to the crude sewage during the ordinary process of filling u
matured coke-bed.
Three strains of "sewage proteus"* were isolated from 1/1000 c.c. of Barking primary B
effluent, (December 6th, 1899). They all grew luxuriantly at blood-heat, formed gas in twentyfour
hours in gelatine shake cultures, liquefied gelatine very rapidly, and were very motile. One
proved virulent when injected into a guinea-pig; another produced a strong local reaction, but
the animal recovered, while the third was apparently non-pathogenic. The latter microbe was
used in the experiment.
700 c.c. of agar were added to about 70 tubes and sterilised. The agar was allowed to
solidify obliquely in the test tubes. The tubes were next inoculated with the "sewage proteus"
and incubated at 20° C.
The growth from each of the tubes was transferred to sterile bouillon (1,500 c.c.), and the
mixture of bacteria and broth used to inoculate the crude sewage as it flowed on to the 13-foot cokebed
at Crossness. This meant the addition of billions of B. proteus to the sewage.
The experiment was carried out in the following way—
On December 19th, 1899, the bacterial mixture (1,500 c.c.) was added to the crude sewage
(about 6,000 gallons) as it flowed on to the coke-beds, the addition being made in small quantities
at a time, so as to cover the whole of the period occupied in filling the bed, namely, about twenty
minutes.
The bed was allowed to remain full for the usual period, namely, three hours, and it was
emptied in the usual manner.
During the process of emptying, and especially during the first flow of the effluent, a
number of vessels in bulk representing approximately thirty gallons—were filled with the effluent
and kept in a warm place.
"When the time arrived for filling the coke-bed again the contents of the vessels were added
to the sewage flowing on to the bed.
Further, when, after the usual period, the bed was again emptied, the vessels were again
filled with the first flow of the effluent.
The above series of operations was carried out until December 22nd, i.e., the crude sewage
(about 6,000 gallons) as it flowed on to the coke-bed was first charged on December 19th with a
mixture of B. proteus and broth (1,500 c.c.) (the microbes being present in number inconceivably
great), and subsequently on six other consecutive occasions with the first flow of the effluent from
the previous emptying of the coke-bed.
Most unfortunately at this period circumstances arose which made it impossible to test the
value of the experiment from the biological point of view. This is greatly to be regretted, as it
would have been of interest to have known if B. proteus was more numerous in the effluent than
normally, and, if so, how long such numerical superiority was maintained. Further, whether its
presence exercised any inhibiting influence on any of the other bacteria usually found in the crude
sewage and effluents.
Mr. Biggs, the chemist in charge at the Southern Outfall Works, kept careful notes during
the above period. He was unable either from the chemical or practical point of view to arrive
at any other conclusion than that the experiment yielded quite negative results. That is to say
the inoculation of the coke-bed with B. proteus did not produce any appreciable alteration in the
effluent either as regards its chemical composition or physical appearance. Nor was the capacity of
the coke-bed increased. The negative results may have been more apparent than real. Possibly
if the effluents had been tested bacteriologically well defined changes might have been observed in
the bacterial composition of the liquid extending over a longer or shorter period. It is conceivabls
also that if B. proteus had been added in still greater amount and for a longer period totally
different results might have been obtained. To overcome the biological equilibrium of a mature
coke-bed of the large capacity of the 13-foot ed at Crossness would naturally call for the addition oft
an enormous number of micro-organisms.
In conclusion, it must be insisted upon that the apparently negative character of this
single experiment in no way suggests the advisability of abandoning a research carried out on the
lines that have been here briefly indicated.
* The morphologicaland biological characters of "sewaje proteu" ave been already described. See
Second Report, V 3, page 36.