The occurrence of sewage fungus in rivers in the United Kingdom

The occurrence of sewage fungus in rivers in the United Kingdom

Water Research Pergamon Press 1971. Vol. 5, pp. 281-290. Printed in Great Britain THE OCCURRENCE OF SEWAGE FUNGUS IN RIVERS IN THE UNITED KINGDOM E. ...

658KB Sizes 0 Downloads 3 Views

Water Research Pergamon Press 1971. Vol. 5, pp. 281-290. Printed in Great Britain

THE OCCURRENCE OF SEWAGE FUNGUS IN RIVERS IN THE UNITED KINGDOM E. J. C. CURTIS and D. W. HARRINGTON Water Pollution Research Laboratory, Stevenage, England (Received 11 December 1970)

Almraet--Scwage-fungus outbreaks are associated with the presence of biodegradable organic matter in water and thus in turn with effluents which discharge such materials. Outbreaks in the United Kingdom are mostly of limited extent except where a fiver has received a succession of polluting effluents. Of the sanitary measurements at present in use BOD and soluble organic carbon provide the most refiable guide to the slime-promoting properties of a water, but they are not infallible since heavy slime outbreaks can be associated with low BOD values and vice versa. Possible reasons for this would include the differing rates at which various biodegradable substances can be utilized and the presence or absence of inhibitors and poisons. Where slime outbreaks are associated with waters of low BOD or organic carbon content the presence of rapidly biodegradable materials not normally present in a good quality sewage effluent (e.g. sugars) has often been noted (AMe~O and ~CK, 1960; PHAUPand GANNON,1967). Adequate secondary biological treatment of an effluent seems in most cases to provide a reliable safeguard against slime outbreaks, but where this is insufficient such treatment followed by a tertiary "polishing" process is likely to be effective. Although the results of the present survey and the 1960 WPRL Questionnaire cannot be directly compared it seems likely that the sewage-fungus situation in rivers in the United Kingdom has not deteriorated in the last decade. OUTe~ o f sewage fungus---cotton-wool-like slimes of attached micro-organisms-are associated with organic pollution of rivers and constitute a troublesome problem when massive growths, which can completely cover submerged surfaces, occur. Sewage-fungus organisms are probably part of the normal microbial flora of all rivers, but form macroscopic shines only where there are significant amounts of biodegradable organic nutrients to serve as growth substrate. While the physiology and nutrition o f the organisms involved--bacteria, fungi, protozoa---are reasonably wellknown from laboratory studies (reviewed by CURTIS, 1969), other aspects of slime growths in rivers--their frequency of occurrence and relation to various types of effluent discharge--are less well documented. In 1960 a questionnaire on sewage fungus was circulated from the Water Pollution Research Laboratory to the River Authorities and Conservancies of England and Wales regarding the extent, suspected causes, and consequences of outbreaks (DEPARTMENTOF SCIENnrlC AND INDUSTRIALRESEARCH, 1962). Information was received on 119 reported outbreaks; chemical data were limited to BOD values of the water at the site of sewage-fungus growth and of the suspected effluent discharge, but neither the sewage-fungus growths nor the sites were examined by staff from the Laboratory. Complaints regarding outbreaks of sewage fungus could be divided into four groups based on (a) appearance and amenity (37.5 per cent o f total), Co) smell, deoxygenation 07"1 per cent), (c) effects o f scoured floes 03"2 per cent), and (d) damage to fish (17.1 per cent). In 14.5 per cent o f cases there was no complaint. To supplement the information obtained from this questionnaire and to note any trends since 1960, a large number of sites of sewage-fungus growth in the United Kingdom were surveyed in 1968 and 1969. This survey, which dealt more fully with 281


E.J.C. CURTISand D. W. ~ T O N

the nature of sewage fungus and the chemical quality of water associated with its growth, forms the main part of the present paper and the findings are compared with data from the 1960 questionnaire (subsequently referred to as the 1960 WPRL Questionnaire). Data from a separate study at this Laboratory of sewage-fungus occurrence and carbohydrate content of water at the site of an outbreak (DEPARTMENT OF SCIENTIFIC AND INDUSTRIAL RESEARCH, 1964) are also included for comparative purposes. METHODS With the assistance of the River Authorities and Conservancies of England and Wales and the River Purification Boards of Scotland, 178 sites of sewage-fungus outbreaks were located and visited during 1968 and 1969. A sample of slime taken at each site was examined for slime-forming organisms, protozoa, and diatoms within 24 h of collection--to minimize changes in composition. The dissolved-oxygen concentration of the water was measured with an EIL dissolved-oxygen meter, water temperature was recorded, and an estimate was made of water flow and velocity. Water samples from each site were collected and preserved by the addition of mercuric chloride solution. After measuring the suspended-solids content, the filtered sample was analysed for soluble organic carbon, soluble phosphate, ammoniacal nitrogen, Kjeldahl nitrogen, oxidized nitrogen, nitrite, and total carbohydrate contents. Measurements of soluble organic carbon were made in preference to B e D determinations because it was thought that the former would provide a more reliable estimate of the amount of organic matter, particularly when the analysis could not be performed immediately, or soon after sampling. B e D values for the period in which the site was visited were obtained, when available, from the River Authorities and Purification Boards who also supplied data regarding effluent discharges suspected of causing the slime outbreak. In order to give an indication of the severity of an outbreak the amount of slime growth was visually assessed/n situ as light, moderate, or heavy and then related to similar assessments of known dry weights of slime growing under experimental conditions at river sites and in the laboratory. "Light growth" represents isolated flocs of sparse growth of less than 5 g dry wt slime m -2. "Heavy growth" refers to a thick slime covering all unscoured surfaces equal to or exceeding 100 g dry wt slime m-Z; entrapped silt and debris can result in. a total dry weight of three times this value. "Moderate growth" represents outbreaks intermediate between these values. No attempt was made to estimate the total biomass of an outbreak. RESULTS

Composition of the sewage-funguscomplex "Sewage fungus" describes the various communities of mostly hcterotrophic organisms forming slime growth in organically polluted rivers. Such slimes comprise a matrix of filamentous organisms (bacteria, fungi, and protozoa) which gives the community cohesion and which contains other, non-filamentous, usually motile, organisms (protozoa, diatoms, and bacteria). An analysis of the interrelationships of the constituent sewage-fungus organisms is still in progress and will be published later, hut the commonest slime-forming organisms (i.e. the filamentous and floc-forming constituents) and their relative abundance are shown in TAnlx I.

The Occurrence of Sewqe Funsus in Rivers


Sphaerotilus natans and zoogloeal bacteria were each present at more than 90 per cent of the sites and each was dominant or co-dominant at more than 50 per cent of the sites. S. natans has been extensively studied (reviewed by Ctntrls, 1969) and has generally been regarded as the most frequent organism to form sewage-fungus slimes. "Zoogloeal bacteria" is an estimate of zoogloeae-forming organisms without attempting further identification; these are difficult to isolate and taxonomically not well understood (ZVIRBULIS and HATT, 1967; C ~ T l ~ E and McCoY, 1967). Our grouping could include Zoogloea ramigera, Pseudomonas spp., and also zoogloeal forms of Sphaerotilus. TABLE ]. OCCURRENCEOF THE COMMONESTSLIME-FORMINGORGANISMS EXPRESSEDAS PERCENTAGEOF 178 s r r ~ EY.AIVaN~ Dominant or co-dominant*

Sphaerotilus natans Zoogloea spp. Beggiatoa alba Carehestum polypinum Geotrichum candidum Flavobacterium sp. Leptomitus lacteus Stigeoclonium tenue Diatoms Fusarlum aquaeduetum

52.1 58"5 6"3 6-3 4.4 3.1 3"1 3.1 4.4 1.9



19.5 30.2 6.9 5.7 2.5 11.3

17.6 3"8 14.5 4"4 0.6 25.8 0.6 1.9 20.7 --



5.7 15.1 --

* This column totals more than 100 per cent because of co-dominance.

Geotrichurn candidum was the commonest true fungus but was dominant in only 4.4 per cent of the sites. This organism has not previously been reported as a major constituent of sewage fungus although it is known to occur in a wide variety of sites including activated sludge (PIPEs, 1967). Flavobacterium sp. is an as-yet unnamed filamentous species composed of long (> 50 pro), thin (about 0.5 pro) rods which are capsulated and contain xanthophyll pigments; these impart a pink or orange colour to colonies in pure culture and sometimes to river slimes. A very similar organism was described in a sewage-fungus slime by O R ~ g o v , G R ~ and Ogsmgov (1966), but no other filamentous Flavobacterium species have been described in this habitat. The composition of sewage-fungus slimes did not appear to be related to the geographical region of the site, similar types of slime were found throughout the various regions sampled. Extent of sewage-fungus outbreaks The outbreaks studied ranged from minor ones involving small streams and effluent channels to extensive growths in large rivers. Data supplied by the River Authorities indicated that 73-5 per cent of outbreaks were less than i mile in length and a further 11 per cent were ~ 1 mile in extent. In 99 per cent of cases the total length ofwatercourse affected was 5 miles or less. Data from the 1960 WPRL Questionnaire confirmed this general pattern (FIG. 1).


E . J . C . CURTISand D. W. HARRINGTON 80 7O 1,1

6o DATA COLL.ECTED IN 19611-I~ .....


m 30 u ~E I




I'0 1"5 2"0 2"5 3"0 3"5 4'0 4"5 I.FJCGTH OF SL.dE OUTlmEAK (ndta)


FIG. 1. Extent of sewage-fungus outbreaks in England and Wales.

Sources of pollution Sewage-fungus outbreaks were found to be associated with a wide variety of effluents, the major groups of which are shown in FI~. 2. The most frequent types of polluting discharge were (a) industrial effluent, (b) domestic sewage, and (c) a combination of these which in total were associated with pollution at 72 per cent of the sites studied. The three groups occurred with approximately equal frequency and resulted in about the same incidence of heavy, moderate and light slime outbreaks (FIG. 2). Effluents from food and drink product industries, in particular distilleries, malting, and fruit and vegetable canning (11.2 per cent of total sites), paper manufacture (10.1 per cent), and textile and dyeing processes (9.6 per cent) were those most frequently associated with outbreaks of sewage fungus. Less common were sites associated in whole or in part with agricultural wastes and tip discharges: the most frequent of these were discharges from piggeries (6.7 per cent of total sites), surface run-off and garage washings (6.7 per cent), and refuse tips (4.5 per cent). These relationships are similar to those found in the 1960 WPRL Questionnaire in which the major groups of effluents associated with sewage-fungus outbreaks were domestic sewage (47.1 per cent), industrial (33.3 per cent), and combinations of these (19.5 per cent). Effluents from food and drink process industries (25-1 per cent) were more commonly associated with outbreaks than those from any other industry; effluents from textile manufacture (7"6 per cent) and paper processing (7,6 per cent) were also frequent.

Water quality at sites of sewage-fungus growth Attempts were made to relate the intensity of sewage-fungus outbreaks to some arbitrary ranges of various chemical parameters of the water. Sewage fungus was found over a wide range of soluble organic carbon contents. While most outbreaks were in waters which could be described as moderately to

The Occ~ie~ce of SewageFungus in Rivers


F~o. 2. Sources of pollution at 178 sites of sewagefungus in the United Kingdom. grossly polluted, containing 6 to 20 mg 1-~, a significant number (21-3 per cent of the total) occurred where there was less than 6 mg l-~. The severity of outbreaks was not closely correlated with content of soluble organic carbon but there was a tendency for heavy growt~ to occur more frequently in the more severely polluted waters. Slimes occurred in waters of a wide range of dissolved-oxygen concentrations mostly at sites with more than 8 mg I -~ and never in those with zero. However, all the concentrations recorded are daytime values and thus do not necessarily represent minimum values. The range of phosphate concentrations at the sites was wide but values were rarely very low (<0.02 mg P 1-1) or very high (>5.0 mg P 1-1). Waters supporting slime growths contained only low or moderate amounts of organic nitrogen (less than 2.1 mg l -~) but a wide range of ammoniacal nitrogen. Suspended solids contents were mostly low or moderate, 88 per cent of the sites having less than 80 mg 1-1. Severity of slime outbreaks was not correlated with any particular concentration range of phosphate, organic and ammoniacal nitrogen, or suspended solids. Although the concentrations of total soluble carbohydrate, measured as glucose equivalents (TABLE 2), ranged from less than 1.1 to 30 mg 1-z, at the majority of sites the water contained between 1.I and 8.0 mg 1-~, with a large proportion in the range 1-1-3.0 mg 1-1. Data collected in the summer of 1963 from 66 rivers chosen randomly and not necessarily containing sewage fungus ( D ~ m ~ oF Sci~STnnC ASD INDUST~L I~S~RCH, 1964) indicated that there was a correlation between carbohydrate content and sewage-fungus occurrence (T~sL~ 2). Unlike the present survey the 1963 sampling included several rivers with relatively high concentrations of carbohydrate. One of these which did not contain sewage fungus had a carbohydrate





Carbohydrate content (mg glucose equiv. 1-1) 0-5

Number of rivers Percentage containing sewage fungus

54 9.2



















content of 46.5 mg 1-~, but several industrial effluents had entered this river and it is possible that materials toxic to the growth of sewage fungus were present. At many of the sites studied in the present survey, pollution was associated with the discharge of several effluents or with the discharge of an effluent to a river of already dubious quality. Where pollution was caused by a single discharge to an unpolluted river, sewage-fungus growth could be directly related to the deterioration in water quality. Several such sites were selected at which the river upstream of the effluent ouffall was unpolluted and did not give rise to slime growths; water samples were taken immediately above the point of discharge in addition to the normal sample at the point of the slime outbreak. Data are shown in TABLE 3 for rivers receiving two of the most common types of polluting effluent--domestic sewage works and paper mills. In both groups of rivers sewage fungus was related to demonstrable deterioration of water quality; domestic sewage effluents increased both organic and inorganic nutrient loadings while the paper mill effluents primarily increased the organic content TAI~E 3. WATm~. Q U A t r ~ ABOW AND BELOW Tim P O L I . t n ~ o EFFLUENT AT SITES OF SEWAGE FUNGUS GROWTH (DATA ARE FROM Sn'ES OF SEWAGE FUNGUS GRow'rH WI-][]~F~POI.J.,UTIONW A S CAUSED BY A SINGLE DISCHARGE TO AN OTHERW]CSECLEAN RIVER)

Domestic sewage effluents (7 sites) Above discharge SolubloorBsnic carbon (mg 1-*) Phosphate (mg P 1-*) Total nitrogen (rag N 1 - 9 Organicnitrogtn




Below discharge 4.2

4- 3.00

Paper mill effluents (5 sites) Above discharge 4-2


Below discharge 22.6


0.05 4- 0.02 3.2 4- 1.45

0.39 -4- 0.38

0.04 4- 0.02

5.0 4- 2.46

3.2 4- 1.82

0.04 =t= 0.03 3.7 -4- 1.80




1.0 =h 0"38

4- 0.18

4- 0.59


(mg N

I-*) A m m o n i a c a l nitrogen (rag N I=*)

0-33 4- 0.05

1-03 4- 0.84

0.90 4-

Soluble carbohydrate




4- 0.18

4- 0.53



0.96 4- 0.59 12.8

4- 18.90

(rag glucom equiv. 1-1)

Dissolved oXypn


4- 2"0

11"2 4-





12"2 =t=0'9

11"2 =[= 0"8


Suspended solids (mg I-*)






The Occurrenceof SewageFungus in Rivers


(soluble organic carbon, organic nitrogen, and soluble carbohydrate). In both groups daytime dissolved-oxygen concentrations remained high and the severity of sewagefungus outbreaks was similar. Outbreaks were most common in waters within the velocity range 16-60 cm s -1, being only rarely found at velocities less than 6 cm s -1 or greater than 60 cm s -t (TABLE4). High velocities prevent slime attachment, while at very low velocities rate of transfer of nutrients may be limiting in all except grossly polluted waters. TAste 4. WAres WLocrrYATsrres oe SeWAOe-VUNOUSotrrat~Ar.s Water velocity(cm s-t) Number of sites Percentageof total

<6 6 3.5





31 17.9

63 36.4

58 33.5

15 8.7

Sewage fungus and'domestic sewage e.~uents Domestic sewage effluents formed the largest single group associated with sewagefungus outbreaks and the data obtained allowed an assessment to be made of the effect of treatment on sewage-fungus growth. Data presented in TABLe 5 indicate that slime outbreaks are largely associated with effluents of poor quality; within this category are effluents from plants which do not provide full secondary biological treatment and also those which do but which are considered by the River Authority to be overloaded (either in terms of flow or of population served with respect to the design capacity). Data for soluble organic carbon content (from collections at the time of the site examination) and B e D (obtained from River Authority analyses during the period of the examination) compared reasonably well at these sites (Fie. 3). DISCUSSION Sewage fungus consists for the most part of an assemblage of heterotrophic microorganisms. Thus the requirements for its growth are supplies of (a) degradable organic nutrients, Co) inorganic nutrients (in particular nitrogen and phosphorus) and, possibly, (c) growth factors, vitamins, etc. Laboratory growth studies indicate that apart from the requirement of Sphaerotilus for trace amounts of vitamin Bt a (MULD~ and VAN VL~, 1962) (usually present in adequate amounts from domestic sewage effluents, agricultural drainage, etc.) a supply of vitamins is probably not necessary for slime growth. It is likely that the supply of inorganic nutrients is only rarely limiting. In experiments at this Laboratory, water with a phosphate content of 0.05 mg P 1-1 has been shown to support heavy slime growths ( C ~ l l s , DELVBS-BROUOHTON and HAg]UNOTON, 1971) and other workers have found that concentrations lower than this are adequate for sewage-fungus growth. Thus TABLE 3, which compares water quality above and below effluents with sewage-fungus outbreaks, indicates that the phosphate content above the effluent discharge (0.04--0-05 mg P 1-1) is ade~luate for heavy slime growth. Although the domestic sewage effluents (T~sLE 3) provide an additional heavy load of phosphorus this is u,likely to have an important bearing on slime growth which seems to be related primarily to an increase in organic content. The small number of sites of sewage-fungus growth at the lowest phosphate content range (<0.02 mg P 1-1) probably reflects the rarity of rivers with phosphate content








® e zo

e +






• •

__ I0



~Z i

w .J



5 ® ÷





I 6


I i i






I t 5O




FIG. 3. Relation of organic carbon content to BOD at sites of sewage-fungusgrowth associated with domestic sewage e~uent. Boxes 1, 2 and 3 indicate areas of different slime frequency (see text). this low rather than phosphate limitation of slime growth. Nitrogen can be utilized in various forms and is usually present in adequate amounts (especially as nitrate) in unpolluted streams. The presence of inorganic nutrients in effluents is thus not likely to influence the development of sewage fungus and its occurrence appears to be most closely correlated with the presence of a source of available carbon. Unfortunately there is at present no method of measuring the proportion of organic matter in this form which is applicable to all situations. B e D is widely used as an indicator of water quality but it is not always a reliable guide. THe ROYALC O ~ O N ON SEWAGeD ~ o s ~ (1912) which classifi~ rivers into five categori~ ranging from very clean to bad, found considerable overlap in the B e D values even between the extreme categories of very deam ( B e D 0~--2.6) and bad (1.3-26.4). Furthermore any delay between sampling and analysis can lead to additional errors. Soluble organic carbon content yields a mote reliable estimate of carbonazeous material but has the disadvantage that it is not necessarily related to the readily biodegradable portion: however, whore pollution was caused by domestic sewage effluents, there was an approximate correlation between the BOD and soluble organic carbon content (Fro. 3). At these sites slime growth

The Occurrence of Sewage Fungus in Rivers




Total no. of occurrences of sewage fungus



























Inadequate or overloaded Adequate secondary Tertiary

Slime growth (percentage of total sites)

BOD at site (rag I -t)



Soluble orllanic carbon at site (rag 1-1) Range Mean


--particularly heavy outbreaks--was especially associated with waters with BOD values of 5--30 (i.e. in the Royal Commission classifications of doubtful and bad) and organic carbon contents of 6-30 (FIG. 3, Box 2). Slime outbreaks in waters with concentrations less than these were not common and were mostly confined to light outbreaks (Box 1). Outbreaks were also rare in waters with BOD or organic carbon contents greater than 30 (Box 3); where slimes were found in such waters they were only of light or moderate extent, possibly indicating the presence of toxic materials in these heavily polluted waters. A wide variety of biodegradable organic compounds is known to support sewagefungus growth (COliTIS, 1969), and effluents likely to contain these compounds were most frequently associated with sewage-fungus outbreaks. The most common indnstrial effluents involved were from food and drink processing industries (waste liquors containing sugars, organic adds, and amino acids), paper manufacture (mostly starch plus some soluble wood polysaecharides), and textile processes (starch, organic acids, esters). Sewage fungus was also widely associated with domestic sewage effluent discharges (F16. 2). There was a strong correlation between inadequacy of treatment and sewage-fungus outbreaks; this was not unexpected bemuse before treatment the soluble portion of domestic sewage contains high concentrations of biodegradable compounds, in particular sugars and organic acids and to a lesser extent proteins, amino acids, and creatinine (P,~aacnm, Vr~Y and BYW^TE~S, 1961). These substances are normally removed during adequate biological treatment but where the treatment process is insufficient or overloaded some is discharged to the river. None of these effluents involved had received "polishing" treatment and in most cases secondary treatment was either absent or inadequate. At the remaining 7 per cent of the sewage works adequate secondary treatment was given to the effluents and slime growths were only light. It also seems that slime outbreaks are associated with poor quality of water in terms of ammoniaeal nitrogen content; a concentration of at least 0.5 nag l-l--above the upper limit of 0.4 mg 1-1 recommended by the World Health Organization as a drinking water standard--was found in 75 per cent of the water samples and 30 per cent had concentrations in excess of 5 mg 1-~. In contrast slime outbreaks do not appear to be associated with poor water quality in terms of suspended solids or of dissolved oxygen. A concentration of suspended solids not exceeding 80 nag 1-1 has been tentatively recommended for maintenance of a fishery by the European Inland WAT~ $16-.--D


E.J.C. CuR3as and D. W. H~P.~OTOt~

Fisheries Advisory Commission; this concentration was exceeded in only 12 per cent of the water samples from sites o f sewage-fungus outbreaks. No standards for oxygen concentrations have yet been made by European countries, but in the United States the National Technical Advisory Committee to the Secretary of the Interior has recommended that to maintain a diverse aquatic community dissolved-oxygen concentrations should generally be above 5 mg l-X. In the present survey a dissolvedoxygen content in excess of 4 mg I - l was found in 89 per cent of water samples. This compares closely with the data obtained in a recent survey of river quality in England and Wales (OweNs, MARIS and ROLLEY, 1970) in which 90 per cent of sampled river miles had a minimum dissolved-oxygen content in excess of 4 mg l-x. Sewage-fungus organisms are known to grow well at low as well as high dissolved-oxygen concentrations, and it seems unlikely that dissolved-oxygen content is an important factor in the occurrence o f slime growth except where there is complete deoxygenation. In addition the data would seem to indicate that in many rivers sewage fungus does not seriously affect the oxygen balance; presumably respiration by sewage fungus is adequately counterbalanced by reaeration. Acknowledfements--Crown copyright. Reproduced by permission of the Controller, H.M. Stationery

Office. We wish to thank River Authorities and River Purification Boards for cooperation and assistance in locating sites. REFERENCES AMBIBlgOH. R. and CtmMAC~J. F. (1960). Factors affecting slime growth in the lower Columbia River and evaluation of some possible control m e a a ~ . Pulp Pap. Can. 61, TT0-T80. C'a~TS~ K. and McCoY E. (1967) Zoofloea ramigera (Itzigsohn), identification and description. Int. J. Syst. Bact. 17, 1-10. CURTmE. J. C. (1969) Sewase fungus: its nature and effects. Water Research 3, 289-311. Cu~11s E, J. C., DEX.VeS-EatOUOWroNJ. and HAmUNOTOND. W. (1971) Sewage fungus: studies of •Sphaerotilus slimes using laboratory recircuiating channels. Water Research 5, 267-279. DEPAgTM~rr OF Scn~ttrtc AND INDUSTPJAt.R m ~ C H (1962) Water Pollution Research 1961. H.M, Stationery ~ , London. D ~ , t ~ ' r ~ r r oF Sc~tcrnnc AND I N D ~ ReS~*,RCH(1964) Water Pollution Research 1963. H.M. Stationery Office, London. MULD~ E. G. and VANV ~ W. Lo (1962) The Sphaerotilus-Leptothrix group. Antonie van Leeuwenhock 28, 236-237. OmailltOD J. G., Gax'~qs B. and ~ O D K. S. (1966) Chemical and physical factors involved in the heterotrophic growth response to organic pollution. Verb. int. Verein theor, angew. Limnol. 16, 906-910. OWENSM., MAres P. J. and ROLLErH. L. J. (1970) River quality--timely reassurance? New Scientist 46, No. 695, 26. P,~n'r~ H. A., VINEYM. and BYWATenSA. (1961) Composition of sewage and sewage effluents. J. Proe. Inst. Sew. Purif. 302-314. PHAtrP J. D. and G~'NON J, (1967) Ecology of Sphaerotilus in an experimental outdoor channel. Water Research 1, 523-541. PrpEs W. O. (1967) Bulking of activated sludge. Adv. appl. Microbiol. 9, 185-234. ROYAL ~ o s ON SewAos D I l ~ a ~ (1912) 8th Report. H,M. Stationery Office, London. Zvn~uLm E. and HATr H, D. (1967) Status of the generic name Zoofloea and its species. Int. J. Syst. Bact. 17, 11-21.