No abstract

Summary. Development of hydrogen sulphidc in sewage, due to growth of sulphate‐reducing bacteria, is favoured by neutral or alkaline conditions and retarded by acidic conditions. Strains able to use lactate, formate, stearate, and cellulose as sources of carbon were detected by the use of appropriate media; no sulphate‐reducing bacteria capable of using tributyrin appeared to be present in the sample of sewage tested. Doses of chlorine much smaller than the chlorine demand delayed the formation of sulphide, sometimes for several days, but eventually the quantity formed was of the same order as that in untreated sewage. Addition to sewage of nitrobenzene, dinitrobenzene, picric acid, or T.N.T., in amounts ranging from 2 to 20 p.p.m. in different tests, delayed development of sulphide during subsequent incubation; sufficiently high concentrations prevented its formation for prolonged periods. Fumarate and nitrate also tended to inhibit formation of sulphide, but the concentrations required (100‐1,000 p.p.m. for nitrate and 1,000–10,000 p.p.m. for fumarate) were much higher than for the above nitro‐compounds. Neither T.N.T. nor picric acid had any marked bactericidal action. It is suggested that these nitro‐compounds retard the lowering of the oxidation‐reduction potential uf the sewage to the region favourable for development of the sulphate‐reducing anaerobes.

Factors affecting the growth ofStreptococcus faecalison glucose-yeast extract-sodium azide agar have been studied. Both the high temperature of incubation used (45° C.) and the presence of azide reduced the proportion of cells able to form colonies, the inhibitory action being much more marked with cultures which had become attenuated, either through age or through prolonged immersion in water, than with comparatively young and vigorous cultures. This inhibitory action was found to be largely overcome if the inoculum was subjected to a preliminary period of ‘resuscitation’, by incubating it with double-strength glucose broth before adding the azide-agar portion of the medium and allowing the mixture to set.Neutral red was so inhibitory to some strains ofStr. faecalisthat it could not be included in the medium. Phosphate, as the potassium salt at a concentration of 0.7%, if autoclaved with the remaining constituents of the medium, exerted a depressing effect on the counts. Added separately it showed no inhibitory action.The spinning-bottle technique (Allenet al.1952) was adapted forStr. faecalis. When used for samples containing a mixed flora the method, described in the Appendix, permitted the growth only ofStr. faecalis.This paper is published by permission of the Department of Scientific and Industrial Research.

Homogenization of activated sludge greatly increased the apparent bacterial count obtained by plating on a solid medium, and for this reason facilitated the isolation of the predominant flora by liberating the bacteria from the interior of the sludge flocs. The ordinary domestic cream-making machine was found to be quite effective for this purpose.The high counts obtained indicated that there are probably several hundred thousand million bacteria per gram of dry matter in activated sludge.Aeration of sewage was found to result in an increase in the total count of bacteria, and a high count was maintained for a considerable time. After aeration for several days fairly rapid and apparently fairly complete separation of the sludge occurred on allowing the liquid to stand, but the supernatant liquor contained large numbers of bacteria. When a fresh quantity of sewage was added to the small quantity of sludge which settled and a second period of aeration commenced, the bacteria were found to associate more quickly with the sludge than they had during the first aeration. Coliform bacteria and spore-forming aerobes formed only a small fraction of the total number of micro-organisms concerned in these changes, and moulds and yeasts never accounted for a large proportion. The predominant bacteria did not produce acid from glucose.Various media were tested in investigating the bacteriology of activated sludge. None showed any superiority to nutrient agar. Coliform bacteria and aerobic spore-formers were encountered in negligible numbers. The morphological and physiological characters of seventy-one strains of the predominant bacteria were determined. The majority were Gram-negative rods with no action on carbohydrates and were members of the generaAchromobacterium, Chromobacterium,andPseudomonas.Those isolated from activated sludge soon after its formation from aerated sewage were found to be non-proteolytic, but a proteolytic flora was established after the sludge had been built up for a period of 4 weeks. The characters of the predominant bacteria indicate that they are water types derived from the water supply or the drainage water which enters sewage. Intestinal bacteria appear to be unimportant.The experiments described in this paper were carried out as part of the programme of the Water Pollution Research Board of the Department of Scientific and Industrial Research, and the results are published by permission of the Department.

SUMlClARY:The growth of Bacterium coli on agar media containing bile salts is conditioned by a number of factors. The proportion of total cells able to grow on a medium not containing bile salts rapidly declines at temperatures of incubation above 43". On a medium containing bile salts and lactose a distinctly inhibitory effect is observed at 37'; with most strains this effect is still more pronounced a t 4 4 ' . Some brands of bile salts are appreciably more inhibitory than others. Inclusion of phosphate in a bile salts medium introduces a markedly inhibitory factor, the severity of which varies with the strain of organism; some strains are virtually unable to grow on such a rnediuni.When a culture of Bact. coli is suspended in water containing only small concentrations of inorganic salts an iticreasing proportion of the population becomes attenuated so that the cells are unable to grow on bile salts lactose agar at 44'. An occasional strain may also exhibit sensitiveness to the presence of neutral red. This attenuating effect may be largely decreased if, before the inoculum is mixed with the bile salts and agar, it is subjected t o a short period of incubation with lactose broth. This treatment has been made the basis of a technique for obtaining a colony count of Bact. coli which is applicable t o polluted waters. (1939). Disadvantages of the method are the large quantity of medium required and the large error involved in the computation of the most probable number. Thus Halvorson & Ziegler (1933) calculated that, with five tubes of medium to each dilution, the count obtained will be between 70% below and Z60y0 above the true value.Enumeration of this organism provides a simple method of following the eff'ect of different treatments of sewage on the numbers of faecal bacteria discharged into a river and of the rate a t which these numbers decrease as the organisms are carried downstream. A colony count of Bact. coli would introduce into sanitary surveys of this character a welcome improvement in accuracy. Clegg & Sherwood (1947) introduced a roll-tube method of counting faecal coli in shellfish in which inocula were mixed with a modified MacConkey agar and the tubes were incubated a t 44". Preliminary tests showed, however, that the ability of cells of Bact. coli to form colonies on bile salts lactose agar