GIBSON DIFFERENTIATION BETWEEN ANIMAL AND VEGETABLE POLLUTION OF WATER 125 The Differentiation between Animal and Vegetable Pollution of Water BY M. GIBSON B.Sc. A.I.C. (Read at the Meeting of the North of England Section October 1941) BUY DENS^ has shown that for waters polluted by vegetable organic substances the oxygen absorption from acid permanganate is greater than from sodium hypochlorite but for waters polluted by animal organic matter the position is reversed. This difference between animal and vegetable pollution has been applied by Dixon and Jenkins,2 with slight modifications of method to the examination of a number of waters showing vegetable and animal pollution. The liquids used in their experiments were made by adding vegetable and animal substances to tap water in such proportions as to give a figure for oxygen absorbed from acid permanganate in 4 hours a t 80" F.of approx. 0.25 part per 100,000 for each solution. Peat leaf mould tea leaves and peas were used as sources of vegetable organic matter and urine and sewage effluent as sources of animal organic matter. The " boiling hypochlorite to boiling perziangana te " ratios for vegetable pollution did not exceed 2 whereas t k s e for animal pollution were between 5 and 8. These results support the opinion expressed by Buydensl that animal and vegetable organic matter in water may be differentiated by their relative action on sodium hypochlorite and yermanganate but whilst Buydens reports the ratio to be >1 for animal pollution and <1 for vegetable pollution Dixon and Jenkins report the ratio to be on the average between 5 and 8 for animal pollution and not exceeding 2 for vegetable pollution.It is not easy to see how the slight variation in analytical methods can explain this difference. TESTS AT BOILING-POINT.-A considerable number of sewage and trade effluents of varying quality have been examined in the laboratory of the Lancashire Rivers Board using the boiling permanganate and sodium hypochlorite methods of Dixon and Jenkins. For 20 samples of sewage effluents (Table I) the average ratio was 2-5 (highest 4.6 lowest 1.1). Two samples of crude sewage (containing no trade waste) gave ratios of 2-2 and 1.7 and three samples of tank effluent (containing no trade waste) gave ratios of 3.3 3 4 and 2-6 re-spectively.Five sample of urine (strong and diluted) gave higher ratios ranging from 5.5 to 8.0. TABLE I-SEWAGE EFFLUENTS OXYGEN ABSORBED N O . 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Treatment Tank and filter . . . . Precipitation tanks and filters . . . . Tanks filters and humus tanks . . Tanks and bio-aeration plant . . . . Tanks filters and humus tanks . . . . . . Precipitation tanks filters and humus tanks . Tanks and filters . . . . . . . . . . Tank and filter . . . . . . . . Tanks filters and humus tanks . . . . . . Septic tanks filters and humus tanks . . . . Tanks and land . . . . . . . . Precipitation tanks filters and humus tanks . . Tanks bio-aeration plant and nitrifying filters . . Tanks filters and humus tanks .. . . . . Tanks filters and humus tanks . . . . Tanks filters and humus tanks . . . . Tanks and land . . . . . . . . . . Septic tanks filters and humus tanks . . . . Tanks and bio-aeration plant . . . . Tanks filters and humus tanks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Boiling KMnO, N j 8 0 52-76 2.36 1.83 1.48 1.43 2.24 5 3 0 20.50 6.34 2-27 1.40 1.70 1 -48 1.70 1.48 1.52 1.17 1.21 1.&0 3.94 Boiling NaOCl 7.64 7.68 3.22 6.83 2-77 6.94 13.38 33-60 19.69 4.66 3.80 4.48 1.69 4.48 2-39 2.74 3.31 2.06 6.56 10.50 hT/10 Average : Ratio NaOCl KMnO, 2.8 3.3 1.8 4.6 1.9 3.1 2.6 1.6 3.1 2-1 2.7 2.6 1.1 2.6 1.6 1.8 2.8 1.7 3-6 2.7 2.5 --Twelve samples of trade waste of vegetable origin (Table 11) have also been examined by the same methods.Two samples of trade waste one from a slaughter house and another from a tannery gave ratios of 6.3 and 3.8 respectively. It will be seen therefore that the results obtained do not agree (except for urine and the slaughterhouse effluent) with those obtained by Dixon and Jenkins but that the ratios show a closer approximation to the results obtained by The average ratio is 1.1 (highest 2.0 lowest 0.37) 126 GIBSON DIFFERENTIATION BETWEEN ANIMAL AND VEGETABLE POLLUTION OF WATER Ruydens. confirmed. The sensitiveness of the action of hypochlorite on animal organic matter is TABLE II-TRADE EFFLUENTS OF VEGETABLE ORIGIN OXYGEN ABSORBED No.1 2 3 4 5 6 7 8 9 10 11 12 Industry Paper manufacture . . . . Paper manufacture . . . . Paper manufacture . . . . Calico printing . . . . Cotton bleaching . . . . Calico printing . . Paper manufacture . . Paper manufacture . . . Cotton bleaching . . . . Dyeing and printing . . Dyeing . . . . . Cotton bleaching and dyeing. . 6 . Boiling KMnO, N/80 9.54 3.60 129.6 31-50 23.0 19.0 5-27 7.89 1930-0 16.0 17-20 10.36 Boiling NaOCl N/10 9-69 5.27 107.7 11-7 23.28 26-1 4.98 10.61 15.52 18-32 20.6 1919.G Ratio NaOCl 1.0 1.5 0.8 0.37 1.0 1.4 0-94 1-3 0-99 0.96 1.1 2.0 KMnO, Remarks Back water Kier liquor Contains starch Contains starch Contains sulphide dyeing waste Back water From sulphite digestion house Starch present Sulphide dyeing Average 1- 1 The next stage in the investigation was to determine to what extent the higher figure for oxygen absorbed for sewages using hypochlorite was due to the greater sezcltiveness of hypochlorite and how much it was due simply to the fact that a stronger solution was being used as the strength of the permanganate used by the previous workers was N/80 or N/100 whilst that of the hvpochlorite was N/IO.The following results (Table 111) have been obtained. TABLE III-SEWAGES OXYGEN ABSORBED Boiling Boiling Boiling N/80 N/10 N/10 KMnO KRlnO NaOCl Ratio Ratio Ratio No.Sample “A” “€3” “C” C/A C/B B/A 1 Final effluent . . ,. 1.17 2-35 3.3 1 2.8 1.4 2.0 2 Final effluent . . . . 1.80 3.86 6.56 3.6 1-8 1-98 3 Tank effluent . . . . 7.00 12.72 18.24 2.6 1.4 1-82 4 Crude sewage . . . . 10-00 15.28 17-36 1.7 1.1 1.53 Ratio C/A is greater than ratio B/A and ratio C/l3 is greater than 1.0. The greater figure for oxygen absorbed is therefore to some extent due to the greater sensitiveness of hypochlorite for animal organic matter and is not solely caused by the higher concentration of the hypochlorite. TESTS AT 80” F.-It was thought that it might be possible to demonstrate this fact by the more usual method for the determination of oxygen absorbed-namely the 4 hours’ oxygen absorbed test a t 80” F. Table IV gives the results obtained for sewage effluents.TABLE IV-SEWAGE EFFLUENTS OXYGEN ABSORBED AT 80” F. N/80 N/80 N/10 KMnO NaOCl KMnO Ratio Ratio N O . Treatment “A” “13” “C” B/A C/A 1 Tanks ,. . . . . . . . . 7-80 12-00 11-24 1.50 1.44 2 Catch pits and filter . . . . . . . . 4-72 10-10 7.00 2-10 1-48 3 Precipitation tanks . . . . 7.24 16-10 9.56 2-20 1.32 4 Precipitation tanks filters and humus tank . . 6.88 9.40 8.84 1.37 1.28 5 Precipitation tanks filters and humus tank . . 5.80 13.00 8.24 2-24 1.42 6 Tanks filters and humus tank . . . . 1.52 2.08 2.08 1-37 1-37 7 Tanks and land . . . . . . . . . . 1-16 5.28 1-54 4-55 1.32 8 Septic tanks filter and humus tank . . . . 1-60 3.36 1-98 2-10 1-24 9 Precipitation tanks filters and humus tank . . 1-34 4-68 1.54 3.50 1.14 double contact filters .. . . . . 2.02 9-72 2-48 4.81 1-20 Average 2-57 1-32 It will be seen that the average K/10 to N/80 permanganate ratio (C/A) for sewages is nearly half the average N/80 sodium hypochlorite to N/80 permanganate ratio (B/A). These results again demonstrate that the higher oxygen absorbed using hypochlorite, is due to the greater sensitiveness of this substance for animal organic matter and is greater than that of permanganate of eight times the normality. 10 Tanks activated sludge and final filters and - _ _ GIBSON DIFFERENTIATION BETWEEN ANIMAL AND VEGETABLE POLLUTION OF WATER 127 A similar series of tests was carried out for trade effluents of vegetable origin and the results are given in Table V. TABLE V-TRADE EFFLUENTS OF VEGETABLE ORIGIN OXYGEN ABSORBED AT SO” F.N/80 N/80 N/10 KMnO NaOCl KMnO Ratio Ratio No. Industry “A” “B” “C” B/A C/A 1 Paper manufacture . . . . . . 6.96 4.04 9.30 0.58 1.34 “2 Paper manufacture . . . . . . . . 1-26 1-28 1-52 1.02 1.21 t 3 Calico printing . . . . . . . . 15-34 7-20 24.80 0.47 1.61 4 Bleaching dyeing and finishing . . . . 2.00 1.86 2-68 0.93 1.34 5 Vegetable gelatine manufacture . . 14.80 6.96 31.20 0.47 2-11 6 Calico printing . . . . . . . . . . 9.30 3.00 21.00 0.32 2.26 7 Paper manufacture . . . . . . 4-76 1-66 5-92 0.35 1-24 8 Paper manufacture (art and chromo department) 0.60 0.60 0.76 1-00 1.27 9 %‘allpaper manufacture . . . . 6-20 2-88 8-16 0-46 1-32 10 Paper manufacture . . . . . . . . 2.02 1.22 2.90 0.60 1.44 11 Cotton bleaching .. . . . . . . . 4.86 2.52 6.92 0.52 1-42 2.28 0.78 1.78 Average 0.63 1.53 - - :12 Cotton bleaching and dyeing . . . . 1.28 1.00 * Green growth present. t Starch present. $. Bleach liquids only. The results given in Tables IV and V indicate that the 4 hours’ test at 80” F. can be used for the differentiation of animal and vegetable pollution. Moreover this method fits into the routine work of a laboratory engaged in water analyses more easily than that suggested by Buydens and by Dixon and Jenkins; it has therefore been adopted. Twenty samples of sewage effluents have been examined and the average ratio was 2.52 (Table VI). TABLE VI-SEWAGE EFFLUENTS OXYGEN ABSORBED AT SO” F. N/80 N/80 KMnO NaOCl Ratio No. Treatment “A” “B” B/A 1 Septic tanks filter and humus tank .. . . . 4.96 15-80 3-2 3 Precipitation tanks filters and humus tanks . . . 0.68 3-12 4-6 5 Tank filters and humus tank . . . . . . . . . 1.48 2.52 1.7 6 Tanks filters and humus tanks . . . . . . . . . . 0.60 0.96 1.6 7 Tank filter and humus tank . . . . 3.60 11-56 3.2 58 Tanks bio-aeration plant and single contact beds . . . . 1.12 4-84 4-3 9 Tanks filters and land . . . . . 3.68 12-40 3.4 2 Tanks filters and humus tanks . . . . . . . . 5-44 11.92 2.2 4 Tanks filters and humus tanks . . . . . . . 1.52 1.54 1.0 10 Precipitation tanks (for laundry waste) tanks filter and humus tank 1.00 2.02 2.0 11 Tanks filter and humus tanks . . . . . . . 6.08 15.48 2.5 14 Precipitation tanks filters and humus tanks . . . 1.40 3.08 2-2 17 Tanks and bio-aeration plant .. . . . . 1-72 6.68 3.9 18 Tanks bio-aeraticn plant and double contact filters . . . . 1.10 2.10 1.9 j19 Tanks and bio-aeration plant . . . . . . 3.34 3.02 0-9 12 Precipitation tanks filter and humus tanks . . . . 0.96 2.46 2.6 13 Septic tanks and filters . . . . . . 1.16 3-06 2.6 15 Tank and double filtration . . . . 0.72 1.44 2-0 16 Precipitation tanks filter and humus tanks . . . . 1.66 4.86 2.9 20 Tank . . . . . . . . 4.24 7.34 1-7 *Average 2.52 I_ * Note average for previous series in Table IV is 2-57. t Sewage contains a high proportion of trade waste of vegetable origin. $ Contact beds inefficient. In three other instances for reasons as yet unexplained the ratios were 8.3 6.4 and 6.0. For 20 samples (Table VII) of trade effluents containing vegetable organic matter An effluent from a tannery gave a ratio of 1.9.The results given in Tables IV to VII indicate that the ratio of oxygen absorbed from AN /80 sodium hypochlorite to oxygen absorbed from acid N/80 potassium permanganate in 4 hours at 80” F. is greater than 1.0 for animal pollution and less than 1.0 for vegetable pollution. the average ratio was 0.52. Uririe gave a ratio of 6.2 128 GIBSON DIFFERENTIATION BETWEEN ANIMAL AND VEGETABLE POLLUTION OF WATER TABLE VII-TRADE EFFLUENTS OF VEGETABLE ORIGIN OXYGEN ABSORBED AT SO” F. S O . Industry N/80 KMnO, “A” 1 J 3 4 5 6 7 8 9 10 ii 12 13 14 15 16 17 18 19 10 Dyeing . . . . . . Wallpaper manufacture . .Paper manufacture . . Dyeing . . . . Cotton bleaching . . * . Cotton bleaching . . . . . . Calico printing . . . . Dyeing . . . . . . Dyeing . . * . Bleaching and printing . . . . Bleaching and printing . . . . Calico printing . . . . Cotton bleaching . . I . . . Calico printing . . . . . . Paper manufacture . . . . Paper manufacture . . Dyeing . . * . . . Cotton bleaching . . * . Paper manufacture . . Cottor bieaching and printing . . 2.84 3.96 2.40 3-80 10.74 3.16 12.96 21.60 85-2 18.0 11.12 12.12 5-80 6.04 10.16 4.54 11.56 7.40 1.28 8.56 N/80 NaOCl Katio “B” B/A 2.10 0.14 1.50 0.38 2.58 1.08 2-68 0.71 3.48 0.32 1-00 0.32 4-52 0.35 14-20 0.66 63.6 0.75 5.8 0.32 4.48 0.40 4.96 0.41 2.04 0.35 4.84 0-80 4-04 0.40 0.82 0.18 4.16 0.36 7.40 1-00 0-64 0.50 2-48 0.29 *Average 0.52 -Remarks Starch present Starch present.Sulphide dye waste present Sulphide dye waste present Starch present Starch present Sulphide dye waste present Free chlorine present * Note that the average for previous series in Table V is 0.63. Experiments were then made to determine first how the ratio of N/80 sodium hypo-chlorite to N/80 permanganate was affected by the quality of the sewage effluents as regards degree of purification and method of purification and secondly the effect of the presence of trade effluents. The following results (Table VIII) have been obtained: TABLE VIII-SEWAGES I N COURSE OF TREATMENT OXYGEN ABSORBED AT 80” F.KO. 1 Bio-filtration plant (entirely domestic sewage) Crude sewage . . Tank effluent . . . . . . . . . . Effluent from filters . . a . ,. Humus tank effluent . . 2 Bio-fcltration plant (sewage contains trade effluent) Crude sewage . . . . . . . . Tank effluent . . Effluent from filters . . . . . . . . Humus tank effluent . . 3 Bio-aeration plant (entirely domestic sewage) Crude sewage . . . . . . . . . . Tank effluent . . . . Final effluent from bio-aeration plant . . Crude sewage . ~ . . . I Tank effluent . . . . . . . . . Final effluent from bio-aeration plant . . 4 Bio-aeration plant (sewage contains trade effluent) Chloride in terms of c1 8.0 8.0 7.4 7.5 15.4 8% 7.6 7.2 7.8 5.8 -31.6 33-3 30.6 N/80 KMnO, “A” 9.3 7.2 7.92 4.8 10.6 6.8 2-76 2-06 6-20 5.4 1.44 15.7 11.68 2.72 N/80 NaOCI “B” 25.9 21-4 8.24 6-16 14-5 13.0 2-46 2.24 11-6 13.0 5.24 19.0 17-6 9-28 Ratio B/A 2.8 3.0 1.04 1.3 1.37 1.91 0.89 1.09 1.87 2.4 3.6 1.1 1-5 3.4 From the above results and from those obtained from the 20 samples of sewage effluents (Table 1’1; compare Nos.5 12 and 20) it is not possible to conclude that the ratio is higher for crude and partially purified sewage than for final effluents. There does not appear to be any relationship between these two factors but the presence of organic trade waste of a vegetable character seems as might be expected to reduce the ratio.This is observed with both bio-filtration and bio-aeration plants a t each stage of the treatment. Also from the figures given in Tables I V VI and VIII it will be observed that the ratio is generally higher for effluents from bio-aeration plants than for those from the bio-filtration plants. This observation might be explained by the fact that few bio-aeration plants contain nitrifying bacteria and thus the proportion of carbonaceous to nitro-genous organic matter is reduced and the hypochlorite to permanganate ratio is therefore increased. A number of miscellaneous samples have also been examined for the presence o GIBSON DtFFERENTIATION BETWEEN ANIMAL AND VEGETABLE POLLUTION O F WATER 129 vegetable and animal pollution :-Interesting results were given by a sewage effluent which contained a high proportion of brewery waste.The effluent was bad and the odour of the sample was very beery : '. 'i: }B/A = 0.83 Oxygen absorbed from N/80 KMnO in 4 hours a t 80" F. (A) . . Oxygen absorbed from N/80 NaOCl in 4 hours a t 80" F. (B) . . Manchester tap water which is known to contain peaty matter gave a ratio of 0.71. Two samples of stream water taken from the Gleaston Beck one above Gleaston Villagc and one below gave the following results on analysis: Source Determination Parts per 100,000 Oxygen absorbed from N/80 KMnO in 4 hours at 80" F. (A) . . village Oxygen absorbed from N / 8 0 NaOCl in 4 hours a t 80" F. (B) . . Free and saline ammonia in terms of N . . . . . . . . 0.009 Albuminoid ammonia in terms of N .. . . . . . . . . 0.002 Oxygen absorbed from N/80 KMnO in 4 hours a t 80" F. (A) . . Free and saline ammonia in terms of N . . . . . . . . 0.130 Albuminoid ammonia in terms of N . . . . . . . . . . 0.062 Dissolved oxygen absorbed in 5 days a t 65" F. . . . . 0.46 ::: }B/A = 0.50 Above Dissolved oxygen absorbed in 6 days a t 65" F. . . . . . . 0.10 Below . . 0.16 lBlh = 3.0 village Oxygen absorbed from N / 8 0 NaOCl in 4 hours a t 80" F. (B) . . . . 0.48 J There is no sewage disposal works a t Gleaston and these results clearly indicate that the brook has received some pollution by sewage matter. Above the village the stream in its pristine condition contains a little natural vegetable matter and consequently the ratio is < I but after receiving the sewage the ratio is increased to 3-0.During the last few weeks the staff of the Lancashire Rivers Board have been trying to locate the source of an intermittent pollution which was the subject of a recent complaifit. I t was necessary in the first place to find out if the pollution was from a sewage disposaI works or from a manufactory. A number of samples of the river water have been examined, and in every instance the hypochlorite-permanganate ratio clearly indicated that the pollution was of vegetable origin and therefore in all probability came from a manu-facturing process. Finally as a result of frequent inspections the pollution has been found to be due to crude piece scouring waste discharged to the stream during the night. The polluting effluents from this factory are derived from the washing of piece goods with soapy liquors.The waste waters do therefore contain large quantities of vegetable organic matter; the hypochlorite to permanganate ratio was found to be 0-69. ANALYTICAL METHODS R. BUYDENS' METHoD.l-oxygen absorbed from acid permanganate.-One hundred ml. of water 5 ml. of sulphuric acid (density 1-3) and 10 ml. of 0.01 N potassium permanganate solution are heated to reach boiling point within 5 minutes and then boiled for 10 minutes. Oxygen absorbed from sodium hypoch1orite.-Twenty ml. of standard 0.1 N sodium hypochlorite solution are added to 100 ml. of water and the mixture is heated to boiling within 5 minutes and boiled for 10 minutes. When cold 2 ml. of a 10 per cent. solution of potassium iodide and 10 ml.of hydrochloric acid (density 1.2) are added and the liquid is titrated with 0.02 N sodium thiosulphate solution a 0-1 per cent. solution of cc-naphtho-flavone in 96 per cent. alcohol being used as indicator in suficient quantity to cause a blue colour. DIXON AND JENKINS' METHOD.2-oXyge?Z absorbed from acid permafzganate.-Ten ml. of A7/80 potassium permanganate solution and 5 ml. of sulphuric acid (1 in 4) are added to 100 ml. of water in a 500-ml. conical flask the mixture is heated to boiling in 5 minutes, and boiling is continued for a further 10 minutes. After cooling potassium iodide is added and the liquid is titrated with N/250 thiosulphate solution starch being used as an indicator. Oxygen absorbed from sodium hypochZorite.-Ten ml. of N/10 sodium hypochlorite solition are added to 100 ml.of the water in a 500-ml. conical flask the mixture is heated to boiling in 5 minutes and boiling is continued for a further 10 minutes. On cooling 2 ml. of 10 per cent. potassium iodide solution and 10 ml. of strong hydrochloric acid are added and the liquid is titrated with N/40 thiosulphate solution a-naphthoflavone (0.1 per cent. solution in alcohol) being used as indicator. (In my experience starch is a satisfactory indicator.) In both titrations the results are expressed in parts of oxygen absorbed per 100,000 The end-point is pink 130 FEARON THE DETECTION OF LACTOSE AND MALTOSE BY MEANS OF METHYLAMINE parts of water. (I have also carried out blank experiments using distilled water in the determination of oxygen absorbed from permanganate and sodium hypochlorite.This avoids the errors which would otherwise be introduced through loss of chlorine on boiling.) NEW METHODS SUGGESTED.-oXJJgen absorbed from acid N/80 permanganate in 4 hours at 80” F.-A suitable quantity of the sample is added to 50 ml. of N/80 permanganate, acidified with 10 ml. of dilute (1 in 4) sulphuric acid (a little permanganate had previously been added to the acid in sufficient quantity to give a faint residual pink colour to oxidise organic impurities) in a stoppered bottle and the mixture is digested for 4 hours at 80” F. Two ml. of 5 per cent. potassium iodide solution are added and the liberated iodine is titrated with N/40 sodium thiosulphate solution starch being used as indicator. Oxygen absorbed from N/80 sodium hypochlorite in 4 hours at 80” F.-A suitable quantity of the sample is added to 50 ml.of N/80 hypochlorite solution and the mixture is digested for 4 hours at 80” F. Two ml. of 10 per cent. potassium iodide solution and 10 ml. of conc. hydrochloric acid are added and the liberated iodine is titrated with N/40 sodium thiosulphate solution starch being used as indicator. Blank experiments were carried out in both determinations distilled water being used in place of the sample. The results are expressed in parts of oxygen absorbed per 100,000 parts of water. In both Dixon and Jenkins’ method and the suggested method the quantity of sample taken for each determination was such as to ensure that approximately 50 per cent. of the oxidising agent remained at the end of the reaction period.CoNcLusIoNs.-The investigations described in this paper support the opinions of Buydens and of Dixon and Jenkins namely that the ratio of hypochlorite to permanganate does provide a method of differentiating between animal and vegetable pollution of water, but it has not been possible to confirm the actual figures given by the last two workers. The only contribution here claimed is that it appears to be possible to use more simple routine analytical methods-and that the ratios obtained are capable of providing the same information. Although as stated in the paper there are one or two abnormal results for which no explanation can as yet be given in no instance has an “animal” ratio been obtained for a water known to be heavily polluted with vegetable organic matter or a “vegetable” ratio obtained for a water known to be heavily. polluted with animal organic matter. For the majority of samples the ratios obtained by the new methods were the ratios that were expected that is <1 for vegetable matter and >1 and sometimes > 2 for animal organic matter. This work was done for the Lancashire Rivers Board as an attempt to assist in the elucidation of rivers pollution problems arising in their laboratory and is published with their permission. The investigation has been carried out under the direction of Dr. G. D. Elsdon F.I.C. Chief Inspector to the Board and I am greatly indebted to him for his guidance and encouragement. REFERENCES 1. 2. Buydens R. “Wa.ter Pollution Research Summary of Current Literature,” 1936,9 (No. 2 ) 51 par. 193. Dixon F. and Jenkins D. C. ANALYST 1939 64 735. THE LANCASHIRE RIVERS BOARD 50 MOSLEY STREET MANCHESTER 2 October 194