TEXAS STATE BOARD OF HEALTH BUREAU OF SANITARY ENGINEERING AUSTIN Presented to the Statistical Division, Surgeon-General's Library, United States Army Washington, D. C. BY The Prudential Insurance Co. of America Newark, New Jersey The Production of Wholesome Water FOR MUNICIPALITIES BY LEWIS O. BERNHAGEN, A.C. Sanitary Engineer OSCAR DAVIS, M. D. State Health Officer UNDER DIRECTION OF V. M. EHLERS, C. E. State Sanitary Engineer PICTURESQUE METHOD OF WATER DISTRIBUTION PUBLICATION NO. 4 920 A561-820-5m-Ijl75 TEXAS STATE BOARD OF HEALTH BUREAU OF SANITARY ENGINEERING AUSTIN The Production of Wholesome Water FOR MUNICIPALITIES BY LEWIS O. BERNHAGEN, A.C. Sanitary Engineer UNDER DIRECTION OF OSCAR DAVIS, M. D. State Health Officer V. M. EHLERS, C. E. State Sanitary Engineer PICTURESQUE METHOD OF WATER DISTRIBUTION PUBLICATION NO. 4 1920 TEXAS STATE BOARD OF HEALTH BUREAU OF SANITARY ENGINEERING OSCAR DAVIS, M. D., President. E. G. EGGERT, Sanitary Engineer. GEO. M. CROOK, Sanitary Engineer. L O. BERNHAGEN, Sanitary Engineer. V. M. EHLERS, C. E., Director. MARTHA D. WILKINSON, Secretary. INTRODUCTION. The purpose of this pamphlet is to present in a condensed form in- formation which will be of assistance to operators of water filtration and sterilization plants where no regular chemist or bacteriologist is in attendance. The first step in the purification of water consists in making analyses of the water to determine some of its characteristics, in order that the right treatment can be applied. It would, therefore, seem proper to begin the subject matter of this pamphlet with a brief description of the necessary tests, both chemical and bacteriological, which might assist the operator to determine the peculiar characteristics of the water he must work with, and also enable him to find out what changes, due to the treatment, have taken place. LIST OF ILLUSTRATIONS. PAGE. Picturesque Method of Water Distribution , Cover Arnold Sterilizer for Sterilizing Culture Media 16 Autoclave for Sterilizing Culture Media 17 Oil Heated Incubator 20 Gas Heated Incubator with Thermo Regulator 20 Development of Bacteria on Culture Plates 21 Illustrating the Principle of Coagulation in Water Treatment 29' Installation for Softening Water by Permutit Process 31 Modified Jewel Gravity Filter ' 35 Continental Air Wash Gravity Filter 34 View in St. Louis, Mo., Plant 35 View of Operating Floor, Minneapolis, Minn., Plant 36. Continental Strainer System with Header and Manifolds 37 Filter Operating Table, Minneapolis Filtration Plant 39 Venturi Meter, Simple Type 43' Venturi Meter, Type "M" Register-Indicator-Recorder 44 Chemical Solution Feeding Device. Constant Level Orifice Box.... 45 Chemical Solution Regulating Valve 4& Proportional Chemical Solution Feeding Device 46 Proportional Chemical Solution Feeding Device ... . 47' Dry Chemical Feeding Machine. ... • 48 Liquefied Chlorine Installation for Sterilizing Water 52' Chlorinating Apparatus Showing Detailed Construction. 55 THE PRODUCTION OF WHOLESOME WATER COLLECTION OF SAMPLES. The bottles for the collection of samples shall have glass stoppers, except when physical, mineral, or microscopical examinations only are to be made. Jugsi or metal containers shall not be used. Sample bottles shall be carefully cleansed each time before using. This may be done by treating with a mixture of sulfuric acid and potas- sium bichromate. A mixture of these chemicals may be kept in stock and used over and over again. The stoppers and necks of the bottles shall be protected from dirt by tying cloth or tinfoil over them. For shipment, bottles shall be packed in cases with a separate com- partment for each bottle. Bottles for bacteriological samples shall be sterilized as explained later under bacteriological analysis. BOTTLES. The greatest possible care should be exercised to obtain a truly aver- age sample of the water to be analyzed. The interval of time elapsing between the collection and the analysis of the sample shall be as short as conditions will permit. ' Samples taken for bacteriological examina- tion should be kept at a temperature near 50 degrees Fahrenheit and not more than twenty-four hours shall elapse before the examination is actually begun. The minimum quantity necessary for making the ordinary physical, chemical, and microscopical analyses of water is two liters or about one- half gallon, for the bacteriological examination, 100 cubic centimeters. SAMPLES. PHYSICAL EXAMINATION. The temperature of the sample, if desired, shall be taken at the time of collection, and shall be expressed preferably in degrees centigrade, to the nearest degree. TEMPERATURE. TURBIDITY. The turbidity of water is due to suspended matter, such as clay, silt, finely divided organic matter, microscopic organisms, and similar material. The standard of turbidity shall be that adopted by the United States Geological Survey, namely, a water which contains 100 parts per mil- lion of silica in such a state of fineness that a bright platinum wire one millimeter in diameter can just be seen when the center of the wire is 100 millimeters below the surface of the water and the eye of the observer is 1.2 meters above the wire, the observation being made in TURBIDITY STANDARD. 4 the middle of the day, in the open air, but not in the sunlight, and in a vessel so large that the sides do not shut out the light so as to influence the results. The turbidity of.such water is arbitrarily fixed at 100 parts per million. PLATINUM WIRE METHOD. For field work the platinum wire method is the most practical one to use. This method requires a rod with a platinum wire (a silver or nickel-plated wire will do for ordinary purposes) 1 mm. in diameter inserted in it about one inch from one end of the rod and projecting from it at a right angle at least 25 mm. Near the other end of the rod, at a distance of 1.2 meters from the platinum wire, a small ring shall be placed directly above the wire through which, with his eye directly above the ring, the observer shall look when making the examination. The rod shall be graduated as follows: The graduation mark of 100 shall be placed on the rod at a distance of 100 mm. from the center of the wire. Other graduations shall be made according to Table 1, which is based on the best obtainable data. TABLE 1.-GRADUATION OF TURBIDITY ROD. Turbidity, P. P. M. Vanishing depth of wire (mm) Turbidity P. P. M. Vanishing depth of wire (mm) 7 ' 1095 70 139 8 971 75 130 9 873 80 122 10 . . 794 85 116 11 729 90 110 12 674 95 105 13 627 100 100 14 587 110 93 15 551 120 86 16 520 130 81 17 493 140 76 18 468 150 72 19 . . 446 160' 68.7 20 426 180 62.4 22 391 200 57.4 24 361 250 49.1 26 336 300 43.2 28 314 350 38.8 30 296 400 35.4 35 257 500 30.9 40 '. 228 600 27.7 45 205 800 23.4 50 187 1000 20.9 55 171 1500 17.1 60.. 158 2000 14.8 65 147 3000 12.1 P. P. M.-Indicates parts per million. (mm)-millemeters. Procedure: Lower the rod vertically into the water as far as the wire can be seen and read the level of the surface of the water on the graduated scale. This reading will be the turbidity of the water. Ob- servations must be made in the open air and at about the middle of the day. The wire must be kept bright and clean. If the observations are made in a container, the wire must not be immersed to a greater depth than the diameter of the vessel. Waters which have a turbidity greater 5 than 500 shall be diluted with clear water before the observations are made. The reading must then be multiplied by the degrees of dilution in order to get the true turbidity. TURBIDIMETRIC METHOD. The simplest and most satisfactory form is the candle turbidimeter. This consists of a graduated glass tube with a flat polished bottom, en- closed in a metal case. This is supported over a standard candle and so arranged that one may look vertically down through the tube at the flame of the candle. The observation is made by pouring the sample of water into the tube until the image of the flame of the candle just disappears from view. Care must be taken not to allow soot or moisture to accumulate on the lower side of the glass bottom of the tube so as to interfere with the accuracy of the observations. In order to insure uni- form results it is necessary to have the distance between the top rim of the candle and the bottom of the tube constant. This distance shall be 7.6 cm., or 3 inches. The observations should be made in a dark- ened room or with a black cloth over the head. Care must be exercised that an average sample of the water is examined. The standard candles may be obtained from any of the larger chem- ical supply houses. The results of turbidity observations shall be expressed in whole num- bers which correspond to parts per million (P. P. M.) of silica and recorded as follows: Turbidity between 1 and 50 recorded to nearest unit - Turbidity between 51 and 100 recorded to nearest unit 5 Turbidity between 101 and 500 recorded to nearest unit 10 Turbidity between 501 and 1000 recorded to nearest unit 50 Turbidity between 1001 and greater recorded to nearest unit 100 COLOB. The color or the true color, of water is the color given to it by sab- stances in solution; that is, it is the color of the water after the sus- pended matter has been removed. In stating results the word "color" shall mean the true color, unless otherwise designated. The "apparent color" shall be considered as including not only the true color, but also any color produced by substances in suspension. It is the color of the original unfiltered sample. Therefore before taking the color of a water it should be filtered through filter paper. There are several methods of obtaining the color of water by the use of standard solutions, but the simplest way is bv comparison with glass disks. Experience has shown that the glass disks used by the United States Geological Survey give results in close agreement with those obtained bv the standard chemical color solutions. The water to be tested is com- pared with glass disks held at the end of metallic tubes through which they are viewed by looking towards a white surface. The glass disks are individually calibrated so that the depth of the color is read directly. 6 The determination of the odor, cold and hot, of samples of surface water is important as the odors are usually indicative of organic growths or sewage contamination or both. The odor of some ground waters is caused by the earthy constituents of the water-bearing strata. The odor of a contaminated well water is often contributory evidence of its pol- lution. Observe and record the odor, both at room temperature and at just below the boiling point, as follows: ODOR. COLD ODOR. Shake the sample violently in one of the collecting bottles, when it is about one-half full and the sample is at room temperature (about 20 degrees C.). Remove the stopper and smell the odor at the mouth of the bottle. HOT ODOR. Pour about 150 c.c. of the' sample into a tall narrow beaker or Erlen- meyer flask. In either case keep the vessel covered and heat almost to boiling. Remove from the heat and allow to cool not exceeding five minutes. Agitate the container by giving it a rotary movement, re- move the cover and smell the odor. Expression of Results. Denote the quality of the odor by one of the following expressions: Aromatic Moldy- Free chlorine Musty. Disagreeable Peaty. Earthy Sweetish. Fishy Hydrogen sulphide. Grassy Vegetable. Express the intensity of the odor by a prefix expressing quality, which mav be defined as follows: Prefix and Definition. None-No odor perceptible. Very faint-An odor that would not be detected ordinarily by the average consumer, but that could be detected in the laboratory by an experienced observer. Faint-An odor that the consumer might detect if his attention were called to it, but that would not attract attention otherwise. Distinct-An odor that would be detected readily and that might cause the water to be regarded with disfavor. Decided-An odor that would force itself upon the attention and that might make the water unpalatable. Very strong-An odor of such intensity that the water would be abso- lutely unfit to drink. (A term to be used only in extreme cases.) CHEMICAL EXAMINATION. HARDNESS. Hardness in water is caused by mineral substances in solution. The principal constituents producing hardness are calcium and magnesium. 7 Carbon dioxide in water increases the solubility of calcium and mag- nesium carbonates, forming bicarbonate. If carbon dioxide is removed from the water by boiling, the bicarbonate is decomposed and calcium and magnesium are partly precipitated. The proportion of calcium or magnesium carbonate that a water can hold in solution depends on the concentration of carbon dioxide, which in turn depends on the temper- ature of the water and the proportion of carbon dioxide in the atmos- phere with which the water has been in contact. Consequently, when the carbon dioxide is removed from the water by boiling or otherwise, the carbonates of calcium and magnesium are partly, but not com- pletely, precipitated, and the hardness of the water is thus diminished •and the water is softened to the extent to which these substances are precipitated. The hardness thus removed is called temporary hardness. The hardness which still remains after boiling is due mainly to calcium and magnesium in equilibrium with sulfate, chloride, and nitrate, and residual carbonate, and is called permanent hardness. Non-carbonate hardness is the hardness caused by sulfates, chlorides, and nitrates of calcium, magnesium, iron, and other metals that form insoluble soaps. DETERMINATION OF TOTAL HARDNESS. (Soap Method.') Total hardness is the temporary hardness plus the permanent hard- ness mentioned above. Procedure: Measure 50 c.e. of the water to be tested into a 250 c.c. bottle, add 50 c.c. of distilled water which has been recently boiled and cooled to room temperature. Add soap solution from a burette, not more than a half of a cubic centimeter at a time, shaking the bottle vigorously after the addition of the soap. Continue the additions of the soap solution until after shaking the bottle a permanent lather re- mains for five minutes over the entire surface of the water in the bottle. Prom the number of c.c. of soap solution used, knowing the strength of the solution, the hardness of the water can be calculated. Example: Used 50 c.c. of the water to be tested. Used 6.1 c.c. of the standard soap solution. Strength of the soap solution-1 c.c. equals 0.00155 gram hard- ness as calcium carbonate. Calculation: 0.00155 times 6.1 equals 0.009455. 0.009455 times 20 equals 0.1891. 0.1891 gram hardness in 1000 c.c. water. 0.1891 times 1000 eouals 189.1. 189.1 gram hardness in 1,000,000 c.e. water, or the water tested has a hardness of 189, P. P. M. Hardness is always reported in parts per million as calcium carbonate. Standard soap solution can be purchased from most chemical supply houses. It should be kept tightly stoppered. A fresh supply should be obtained at least every six months. 8 In making the above test, if the water requires more than 7.0 c.c. of the soap solution per 50 c.c. of the water, use 25 c.c. water, adding 75 c.c. of distilled water. If 25 c.c. of the water are used, multiply by 40 instead of 20 in the calculations. TEMPORARY HARDNESS. (Titration with Acid.) Determine the alkalinity in presence of methyl orange (see under determination of alkalinity) in the original sample and also in the sample after boiling, cooling, restoring to the original volume with boiled distilled water, and filtering. The difference between the two, if any, is the temporary hardness. This is the most accurate way of determining the temporary hardness of ordinary waters. Iron bicar- bonate is included as a part of the temporary hardness. PERMANENT HARDNESS. Permanent hardness may be calculated for waters which are soft or moderately hard in a fairly satisfactory manner by subtracting the total alkalinity from the total hardness as obtained by the soap method. This method is not very accurate for waters that are unusually hard or for those that contain much magnesium. ALKALINITY. The alkalinity of a natural water represents its content of carbonate, bicarbonate, borate, silicate, phosphate, and hydroxide. Alkalinity is determined by neutralization with standard sulfuric acid in the pres- ence of phenolphthalein and either methyl orange, erythrosine, or lac- moid as indicators. Methyl orange may be used except in waters con- taining aluminum sulfate or iron sulfate. The alkalinity of carbonates in the presence of phenolphthalein is different from that in the presence of methyl orange, partly because of loss of carbon dioxide and partly because of defects in phenolphthalein as an indicator under such con- ditions. Add four drops of phenolphthalein indicator to 50 or 100 c.c. of the sample in a white porcelain casserole or in a beaker over a white back- ground. If the solution becomes colored, hydroxide or normal car- bonate is present. Add N/50 sulfuric acid slowly from a burette, stir- ring the solution constantly while adding the acid, until the coloration disappears. The phenolphthalein alkalinity in parts per million of calcium car- bonate is equal to the number of cubic centimeters of N/50 sulfuric acid used multiplied by 20 if 50 c.c. of the sample was used, or by 10 if 100 c.c. was used. PROCEDURE WITH PHENOLPHTHALEIN. PROCEDURE WITH METHYL ORANGE. Add two drops of methyl orange indicator to 50 or ^.00 c.c. of the sample in a beaker over a white background. If the solution becomes 9 yellow, hydroxide, normal carbonate, or bicarbonate is present. Add N/50 sulfuric acid from a burette until the faintest pink coloration appears. The methyl orange alkalinity in parts per million of calcium car- bonate is equal to the total number of cubic centimeters of N/50 sul- furic acid used multiplied by 20 if 50 c.c. of the sample was used, or by 10 if 100 c.c. was used. PROCEDURE WITH ERYTHROSINE. Put 50 or 100 c.c. of the sample into a 250 c.c. clear glass-stoppered bottle, add 5 c.c. of neutral chloroform and 1 c.c. of erythrosine in- dicator. If the chloroform becomes rose colored on shaking, hydroxide, bicarbonate, or normal carbonate is present. Add N/50 sulfuric acid, in not more than one-half c.c. amounts, from a burette. Shake the bottle thoroughly after each addition of the acid. Repeat the additions of the acid until the chloroform becomes colorless. Toward the end of the titration, when the chloroform is beginning to fade in color, less than one-half c.c. acid should be ad led at a time, otherwise the end- point may be passed unnoticed. A white piece of paper held behind the bottle aids in detecting traces of color. The chloroform to be used in making this test should be of a good quality. The kind used for anesthesia will be satisfactory. This method for determining alkalinity is the best for waters low in alkalinity that have been treated with alum or iron sulfate. The alkalinity of a natural water determined by either the methyl orange or erythrosine methods should be the same. It is commonly called "total alkalinity." INTERPRETATION OF RESULTS. Bicarbonate (HCO3). Bicarbonate is present if the alkalinity to phenolphthalein is less than one-half the alkalinity, to methyl orange or erythrosine. The al- kalinity to methyl orange or erythrosine is due entirely to bicarbonate if there is no phenolphthalein alkalinity. If there is phenolphthalein alkalinity the bicarbonate, in terms of calcium carbonate, is equal to the methyl orange or erythrosine alkalinity minus twice the phenolphtha- lein alkalinity. Normal carbonate is present if the alkalinity to phenolphthalein is greater than zero but less than the alkalinity to methyl orange, or erythrosine. If the phenolphthalein alkalinity is exactly equal to one- half the methyl orange or erythrosine alkalinity, the alkalinity is due entirely to normal carbonate. If the phenolphthalein alkalinity is less than half the methyl orange or erythrosine alkalinity, normal carbonate expressed in terms of calcium carbonate is equal to twice the phenolph- thalein alkalinity. If the phenolphthalein alkalinity is greater than one-half the methyl orange or erythrosine alkalinity the normal car- bonate is equal to twice the difference between the methyl orange or erythrosine alkalinity and the phenolphthalein alkalinity, j Normal Carbonate (CO3). 10 Hydroxide, Radical (OH). If hydroxide, or caustic alkalinity, is present the alkalinity to phe- nolphthalein is greater than one-half the alkalinity to methyl orange or erythrosine; the alkalinity is due entirely to hydroxide if the phenolph- thalein alkalinity is equal to the methyl orange or erythrosine alkalinity. If the phenolphthalein alkalinity is more than half and less than all the methyl orange or erythrosine alkalinity, hydroxide, expressed in terms of calcium carbonate, is equal to twice the phenolphthalein alka- linity minus the methyl orange or erythrosine alkalinity. In the operation of filtration plants, where lime or soda ash are used, care should be taken at all times that the phenolphthalein alkalinity does not exceed one-half the alkalinity obtained when using erythrosine or methyl orange indicators. If the phenolphthalein alkalinity exceed one-half of the erythrosine or methyl orange alkalinity, the water is "caustic." This condition should be carefully avoided, as caustic water will produce all sorts of trouble in the filter beds and also in the mains. 1. Standard sulfuric acid. A N/50 solution. 2. Phenolphthalein indicator. Dissolve five grams of a pure grade of phenolphthalein in one liter of 50 per cent alcohol. The alcohol should be diluted with freshly distilled water. (Make up one-fifth of the above quantity.) 3. Methyl orange indicator. Dissolve 0.25 gram of a good grade of methyl orange in 500 c.c. distilled water. Keep the solution in a dark bottle. 4. Erythrosine indicator. Dissolve 0.25 pram of erythrosine (the sodium salt) in 500 c.c. of freshly distilled water. The standard acid must be purchased from chemical supply houses. The indicators may be put up by anv reliable druggist. Standard N/50 sulfuric acid is of such strength that if 100 c. c. of water are being tested, the number of c.c. of acid used multiplied by 10 will be the a-RcaHtiity. CHEMICAL REAGENTS REQUIRED TOR ALKALINITY TESTS. TOTAL ACIDITY. Add four drops of phenolphthalein indicator to 100 c.c. of the sample in a beaker over a white background. Add N/50 sodium carbonate until the solution turns pink. The number of cubic centimeters of N/50 sodium carbonate used multiplied by 10 will be the total acidity in parts per million. The standard N/50 sodium carbonate solution must be purchased from chemical supply houses. The indicator is the same as that used in the determination of caustic alkalinity. Acidity in natural water is comparatively rare, except that caused by the presence of free carbonic acid. Acidity may also be caused by mineral acids, or some of their compounds, especially those of iron and aluminum. Free Carlon Dioxide (COJ. Pour 100 c.c. of the sample of water to be tested into a tall narrow beaker. Add about ten drops of phenolphthalein indicator, and run in 11 rapidly from a burette, N/22 sodium carbonate solution, stirring gently, until a faint but permanent pink is produced. The number* of cubic centimeters of standard N/22 sodium carbonate solution used, multi- plied by 10 will equal the parts per million of free carbon dioxide. Free carbon dioxide escapes verv readily-from the water, so special care must be exercised in taking and caring for the sample. This is particularly true of waters that are heavily charged with the gas. When taking the sample, the bottle should be completely filled, so as to leave no air space under the stopper. If it is necessary to store the samples, they should be kept on ice. The free carbon dioxide test is of value in plants where considerable alum must be used. If the gas is present in considerable quantity, over eight or ten parts per million, it might in time have some effect on the mains of the waterworks system. EREE MINERAL ACIDS. Add two drops of methyl orange indicator to 100 c.c. of the water in a glass beaker. Set the beaker on a white piece of paper. Add from a burette N/50 sodium carbonate until the pink coloration of the solu- tion disappears. The number of cubic centimeters of N/50 sodium car- bonate solution used multiplied by 10 will give the acidity of the water in parts per million. If the water is not pink after the addition of the methyl orange in- dicator, no free mineral acid is present. CHLORIDE. Chloride in water has its origin in common salt, from mineral de- posits in the earth, from ocean vapors carried inland by the wind, or from polluting materials like sewage and trade wastes, which contain the salt used in the household and in manufacture. Comparison of the chloride content of a water with that of other waters in the vicinity known to be polluted, frequently affords useful information as to its sanitary quality. If, however, the chloride normally exceeds twenty parts per million because of chloride bearing mineral deposits the chloride content of a water has little sanitary significance. METHOD OF DETERMINATION. Pour 50 c.c. of the water to be tested into a white porcelain bowl or into a beaker placed over a white background. Add 1 c.c. of potassium chromate indicator. Run in from a burette, standard silver nitrate solution until a faint reddish coloration begins to appear. The end- point may be detected more readily if the water in the bowl which is being titrated is compared with water in another bowl to which has been added an equal quantity of indicator. If the amount of chloride is quite high, use a smaller amount of water, and add distilled water to make the volume up to about 50 c.c. Example: Used 50 c.c. of the water to be tested. Used 3.1 c.c. of the standard silver nitrate solution. Strength of the silver nitrate solution-1 c.c. is equivalent to 0.0005 gram of chloride. 12 Calculation: 0.0005 times 3.1 equals 0.00155. 0.00155 times 20 equals 0.03100. 0.0310 gram chloride in 1000 c.c. water. 0.0310 times 1000 equals 31.0. 31.0 gram chloride in 1,000,000 c.c. water or the water tested contains 31 P. P. M. chloride. Both the standard silver nitrate solution and the indicator solution must be ordered from chemical establishments which make up standard solutions. RESIDUAL CHLORINE. In waters that have been treated with calcium hypochlorite or liquid chlorine it is frequently advisable to ascertain the presence or absence of chlorine. THE ORTHO-TOLIDIN TEST FOR RESIDUAL CHLORINE. Procedure: Add 1 c.c. of the tolidin reagent to about 200 c.c. of the water in a tall beaker. Allow it to stand for about three minutes. Small amounts of free or residual chlorine produce a yellow and larger amounts an orange color. THE STARCH-IODINE TEST FOR RESIDUAL CHLORINE. To about 200 c.c. of the water in a tall beaker add about one tea- spoonful of the starch-iodide test solution. Allow it to stand several minutes. Small amounts of chlorine produces a faint blue and larger amounts a dark blue color in the water. The ortho-tolidin test is very delicate. With this method it is pos- sible to detect a very minute trace of free chlorine, or one part in 200,000,000 parts of water. When making this test in connection with the operation of a water sterilization plant, the test should be applied directly after the chlorine is added to the water. In either test, the reagent should be thoroughly mixed with the sample of water. Reagents.-Ortho-tolidin solution. Dissolve 0.5 gram of the purified ortho-tolidin in 500 c.c. of 10 per cent hydrochloric acid. Any drug- gist can prepare this reagent. Starch-iodide solution. To 100 c.c. of cold water add about one-half a teaspoonful of corn starch, mix thoroughly with the water. Heat to boiling, add about 1 gram of potassium iodide. Put in a bottle and keep stoppered. This solution can be prepared by anyone. It should be mixed fresh every three or four days. 2 burettes, capacity 50 c.c. 2 burettes, capacity 25 c.c. 6 beakers, tall form, capacity about 250 c.c, 6 Erlenmeyer flasks, capacity about 250 c.c 3 casseroles, capacity about 250 c.c. 1 graduated cylinder, capacity 25 c.c. 1 graduated cylinder, capacity 50 c.c. APPARATUS REQUIRED FOR THE CHEMICAL ANALYSES OUTLINED ABOVE. 13 1 graduated cylinder, capacity 100 c.c. 1 volumetric flask, capacity 500 c.c. 2 burette stands. 6 glass-stoppered bottles, capacity 250 c.c. 6 glass-stoppered bottles, capacity one-half gallon. 6 glass-stoppered bottles, capacity one gallon. 6 two-oz. dropping bottles for indicators. 1 thermometer graduated in degrees to 150 degrees C. 2 glass funnels, diameter 3| inches. 2 glass funnels, 'diameter 4 inches. 2 packages qualitative filter paper, circular, diameter 6 inches. 2 packages qualitative filler paper, circular, diameter 7 inches. 6 1 c.c. pipettes. 6 6-inch stirring rods. 1 turbidimeter, U. S. Geological Survey type, complete for both color and turbidity, or 1 U. S. Geological Survey color outfit, separate, and 1 Jackson turbidimeter. 1 one-burner gasoline stove. 1 thermometer graduated in degrees to 300 degrees C. CHEMICALS REQUIRED. 1 litre standard soap solution. 2 litres standard N/50 sulfuric acid. 1 litre standard N/50 sodium carbonate solution. 1 litre standard N/22 sodium carbonate solution. •J litre silver nitrate solution (for chloride test). 8-oz. methyl orange indicator. 8-oz. phenolphthalein indicator. 8-oz. erythrosine indicator. 8-oz. potassium chromate indicator. 8-oz. ortho-tolidin test solution. 2 lbs. chloroform. 6 lbs. commercial sulfuric acid. 1 lb. potassium bichromate. 4 ozs. potassium iodide. 1 lb. corn starch. FIRMS DEALING IN CHEMICAL APPARATUS, STANDARD SOLUTIONS AND CHEMICALS. E. H. Sargent & Co., 155 E. Superior St., Chicago, Ill. A. H. Thomas Company, West Washington Square, Philadelphia, Pa. A. Daigger & Co., 54 West Kinnie St., Chicago, Ill. Henry Heil Chemical Co., St. Louis, Mo. J. T. Baker Chemical Co., Phillipsburg, N. J. Wallace & Tiernan Co., Inc., 137 Centre St., New York City, N. Y. 14 BACTERIA. Bacteria are the lowest form of plant life at present recognized. They are very widely distributed throughout nature, and are found in all nat- ural waters. They are more numerous in surface than in ground waters. Most all of the species are not only harmless, but are absolutely neces- sary in the development of animals and plants. Without bacteria the present forms of plant and animal life could not exist. There are, however, several species of water borne bacteria that are harmful; these are generally spoken of as the pathogenic species. Sur- face water may be infected by the organisms either by means of domestic sewage or manufacturing wastes. It is, of course, only with the water borne types of bacteria that the waterworks men need concern themselves. Among the most prominent and at the same time the most dangerous of these organisms are the bacillus typhosus, which cause typhoid fever, the bacillus dysenteriae, which produces dysentery, and the spirillum of Asiatic cholera. It is extremely difficult to isolate these particular organisms when, as they are, associated with a great number of other bacteria which com- plicate the determination and cloud the result. These three different species of bacteria are always associated with a group of organisms which have the faculty of fermenting the sugar lactose with the formation of gas. These gas formers are of two types, those forming spores and non- spore formers. In this small pamphlet it will not be possible to ex- plain the complicated procedure necessary to separate these two types, but both will be considered together as one class, all will be designated B. coli, as most of them are of this type. It has been discovered as the result of a great many experiments, that whenever either of the three pathogenic species mentioned above are present in a water, a very great number, comparatively speaking, of the gas producing types are present. In no case has it ever been demonstrated that the three species of path- ogenic bacteria have been found in water where no gas formers were found. With these facts in mind, certain standards of purity for water have been worked out. In other words, the presence or absence of B. col? determines the pollution or purity of the water. The standard of purity for an ordinary water as adopted by the United States Treas- ury Department is in part as follows: "1. The total number of bacteria developing on standard agar plates, incubated 24 hours at 37 degrees C., shall not exceed 100 per cubic centi- meter; provided, that the estimate shall be made from not less than two plates, showing such numbers and distribution of colonies to indicate that the estimate is reliable and correct. "2. Not more than one out of five 10 c.c. portions of any sample examined shall show the presence of organisms of the bacillus coli group when tested as follows: G(a) Five 10 c.c. portions of each sample tested shall be planted, each in a fermentation tube containing not less than 30 c.c. of lactose peptone broth. These shall be inoculated 48 hours at 37 degrees C., and observed to note gas formation." If there is more than 10 per cent gas in the upper part of the in- verted inner tube, that is if the culture media in more than one-tenth of the length of the tube has been displaced by gas, the sample of water shows evidence of a positive presumptive test for B. coli. 15 In a small laboratory where no equipment is provided for further isolation tesfs, this positive presumptive test will have to be considered as indicating the presence of B. coli. Lactose bile broth has been found to inhibit, or keep from growing, certain forms of bacteria that form gas in. lactose peptone broth and do not belong to the B. coli group proper. For this reason the use of lactose bile broth in preference to lactose peptone broth is advised. METHODS FOE BACTERIOLOGICAL EXAMINATIONS OF WATER. APPARATUS REQUIRED. Sample Bottles.-Ground glass-stoppered bottles of four or eight-ounce capacity are recommended. The four-ounce size is large enough for ordinary purposes. The bottles should not be filled more than three- quarters full with the water to be tested. They should be protected by being wrapped in paper, or their necks covered with tinfoil, and should be placed in proper containers for transportation. Pipettes.-Two sizes of pipettes are necessary, the 1 c.c. and 10 c.c. size. Dilution Bottles.-Bottles for making dilutions should have close fit- ting glass stoppers and be of such capacity as to hold at least twice the volume of water actually used. Petri Dishes.-Standard Petri dishes are ten centimeters in diameter. They may be used either with glass or porous tops. The bottoms of the dishes should be flat and made of clear glass free from bubbles. Fermentation Tubes.-Any type of fermentation tube may be used. The most convenient form is the Jackson tube. This tube consists of a homeopathic vial inverted in a test tube, or a small test tube inverted in a larger tube. Sterilization Oven.-The oven may be heated either by gas, oil or electricity. It should be so constructed that the heat will be distributed evenly, and should be provided with a thermometer. MATERIALS FOR CULTURE MEDIA. Distilled water should be used in the preparation of all culture media and reagents. The Digestive Ferments Company's products are recommended. Lac- tose Bile, Lactose Broth, Nutrient Agar and Endo's Agar are the four media necessary. These are put up in powder form and need only to be mixed and dissolved in distilled water tubed and sterilized. They may be ordered through any druggist. STERILIZATION OF GLASSWARE. The sample bottles should be thoroughly cleaned and rinsed with dis- tilled water. A small piece of paper should be put into the necks of the bottles with the stoppers to prevent the stoppers from setting after they are sterilized. Muslin or tinfoil should be wrapped around the tops and necks of the bottles before they are put in the sterilizer. The pipettes should be put in metal containers, made for that purpose. Dilution bottles should be handled the same as the sample bottles. Petri dishes should be very carefully cleaned and rinsed with distilled water, thoroughly dried with a linen cloth and put in a culture dish holder before sterilizing. 16 • All glassware must be sterilized for at least one and one-half hours at a temperature of 170 degrees centigrade in the dry oven. PREPARATION AND STERILIZATION OF CULTURE MEDIA. Dissolve 70 grams of the powdered Lactose Peptone Bile in 1000 c.c. of water. In order to thoroughly dissolve this material put the powdered medium in a mortar and rub it up with a small amount of water, grad- ually forming a paste. Add more water to this paste' and continue to dissolve the material by carefully stirring it. As the pasty material dissolves, pour the solution into a clean vessel and add more water until the whole is in solution. It should amount in quantity to 1000 c.c. To Fill Test Tubes.-Place the small test tubes mouth downwards in- side the large tubes. Pour about three inches of the bile into each of the larger test tubes. Put a wad of cotton into the mouth of the big ^ubes sufficient in amount to keep all dust and dirt away from the media. The inner tubes will not fill up with the bile until the tubes containing the bile are sterilized. STERILIZATION. There are two methods of sterilization: the first intermittent, and the second by means of an autoclave. With intermittent sterilization, the tubes containing the bile and stoppered with cotton are placed in an Arnold Sterilizer for Sterilizing in Free Steam. Arnold sterilizer, the bottom of which is filled with water. The steril- izer is'then heated to a boiling temperature so that a stream of live steam passes up over the tubes for a period of at least thirty minutes. After cooling, the tubes should be removed from the sterilizer and stored, preferably in a refrigerator. This process is repeated on three (Courtesy of E. H. Sargent & Co.) 17 successive days, after which time the contents and the tubes will be thoroughly sterilized. The second method of sterilization is by means of an autoclave. When this method is used the tubes and contents are subjected to live steam under pressure. The contents of the autoclave are kept at a pressure of 15 pounds for fifteen minutes, after which time they will be sterile. Autoclave for Sterilizing Culture Media and Bacteriological Apparatus Under Steam Pressure. (Courtesy of E. H. Sargent & Co.) Time can be saved by using this process, but necessitates the purchase of a somewhat expensive piece of apparatus. Satisfactory sterilization can be accomplished by the use of the cheaper Arnold sterilizer. It will be found, as a result of the process of sterilization, that the inner tubes are completely filled with bile, although there might occa- sionally be a bubble of air at the very top of the inner tube. If the 18 media has been properly sterilized, this will not interfere with the de- termination. NUTRIENT AGAR. Eighteen grams of the powdered nutrient agar are mixed with a small portion of a liter of water, thoroughly stirred and then added to the balance of the liter of water. The whole is poured into a double boiler and heated until all is dissolved. Stirring will assist in the solu- tion. After the solution is complete, make up to the original volume by the addition of distilled water. Agar begins to solidify at 38 degrees C. or about 100 degrees F. For that reason it is necessary to tube it while still hot. Ten c.c. of the media should be put in each tube. This can be conveniently accom- plished by measuring 10 c.c. in one of the tubes and then by observing the height of the media in the tube, fill the other tubes to an equal height. Sterilization of the agar is done in the same manner as that of the bile media. The completed and sterilized media should be stored in a refrigerator. ENDO'S AGAR. Endo's agar is not used as a regular or routine media. It is used for the differentiation between B. coli and other members of the group of bacteria that ferment Lactose Peptone Bile. It need only be pre- pared where the operator wishes to investigate a little more closely into the various types of bacteria that he may have to deal with. It is also a little more difficult to prepare. In the preparation of Endo's media, 55.5 grams of the powdered agar are dissolved in 1000 c.c. of distilled water in a double boiler (the type used for the preparation of oatmeal is satisfactory). Care should be used not to overheat the media during the process of dissolving the agar. After solution is complete, make up the loss in volume due to evaporation, with distilled water. It must be tubed while hot. Steril- ize in an autoclave if possible, otherwise in an Arnold sterilizer in a manner similar to that employed in the preparation of the nutrient agar. Between sterilizations the media should be stored in a refrigerator. Only small portions should be prepared at a time, as the media should be fresh when used. PRESUMPTIVE TEST FOR«B. COLL Samples.-Great care must be exercised in taking samples. If the specimens are carelessly taken and have been contaminated, the work following, no matter how conscientiously performed, will be useless. The bottles used in taking the samples must have been thoroughly sterilized. If the water is taken from a faucet, the mouth of the faucet should be flamed with a handful of burning paper or similar material. The water should then be turned on and allowed to flow for three or four minutes, after which the sample should be taken directly from the flowing stream. The bottle should be held in such a wav that no backwash of water that has touched the hands will get in the bottle. The stopper, after it has been removed'from the bottle, should be held in the hand with the cloth about it until the required amount of the water has been run into the METHOD OE PROCEDURE. 19 bottle. The stopper should then be replaced in the bottle and the cloth tied in place. Do not lay the stopper down, as it may become contam- inated, and on being replaced in the bottle would transfer the contami- nation to the water in the bottle. If the water must be taken from a basin, reservoir or lake, the stopper should be carefully removed, the bottle grasped near the bottom with the mouth of the bottle to the front, pass the bottle through the water with a continuous forward move- ment. By regulating the speed of the movement, the requisite amount of water can be admitted into the bottle. Inoculation.-In routine work 1 c.c. and 1/10 c.c. of the untreated water should be used to inoculate the tubes. If the water is not seri- ously polluted, 10 c.c. and 1 c.c. volumes would be proper. With the treated water 10 c.c. should always be used. The tubes should be large enough to hold 30 c.c. of media plus 10 c.c. of water. When inoculating with 10 c.c. of water at least 30 c.c. of bile broth should be in the tubes. It is assumed that Jackson tubes are used. For the smaller volumes a smaller set of tubes may be employed and considerable media saved by so doing. To make the inoculation before taking any water from the bottles they should be shaken vigorously at least twenty-five times. Remove the stopper from the bottle (lay it aside if the water is not to be saved), take the cotton plug from the test tube with the left hand, holding it between the fingers in such a way that it will be on the back side of the hand. Flame the lip of the tube by passing it back and forth through the flame of a spirit lamp for a few seconds. Remove the pipette from the pipette case with the right hand, using care to touch only the one which is to be used. Put the pipette in the water and with the mouth carefully draw up the water into the pipette until it is slightly, above the graduated mark. Quickly remove the end of the pipette from the mouth, at the same time covering it with the index finger of the right hand. By slightly releasing the pressure of the finger allow the water to flow back to the graduation mark. Insert the tip of the pipette into the test tube and allow the water to flow into the tube. Carefully replace the cotton plug. Agitate the tube slightly so as to mix the contents, and place it in the 37 degree incubator. This method of procedure applies to any of the volumes used for inoculation. If after the end of forty-eight hours samples show more than 10 per cent gas in the inverted tube, the test is known as the positive pre- sumptive test for B. coli. TOTAL NUMBER OE BACTERIA. Plating.-The same set of samples used for the B. coli test may he used for the determination of the total number of bacteria. It is best to do the plating at the same time that the bile tubes are inoculated. The requisite number of tubes of agar are melted by heating in boil- ing water for five or six minutes. This can be conveniently accom- plished by using a water-bath with a rack to hold the tubes upright. After the' agar is melted, remove the water-bath from the fire and allow to cool. Arrange the required number of Petri dishes on a table having a smooth level surface. Number them to correspond with the sample and with the tubes used for the B. coli test. Remove the water from the sample bottle with a 1 c.c. pipette, using the same care exercised in inoculating the bile tubes. Tilt the covers just enough to admit the 20 pipette and allow the water to run into the dish. It is best to use a different sterile pipette for each inoculation or plating, even if both are carried out at the same time, as they properly should be. If water is accidentally drawn up into the mouth, set aside the sample and take a fresh one in another sterile bottle. (Courtesy of E. H. Sargent & Co.) Oil Heated Incubator. Gas Heated Incubator with Thermo Regulator. (Courtesy of E. H. Sargent & Co.) After the samples have been measured into the various dishes, cool the water in which are the agar tubes, down to 40 degrees C. Allow the agar to remain at this temperature for about four minutes and then 21 pour into the dishes. The procedure is as follows: remove the cotton plugs from the agar tubes as they are taken from the bath, sterilize the lips of the tubes by passing through the flame of a spirit lamp, pour the agar into the dishes, raising the covers only enough to admit the agar. Replace the cover immediately and give the plate a rotary motion, keeping it flat on the table while so doing. This manipulation is neces- sary in or.der to thoroughly mix the agar and water and also to dis- tribute the media evenly over the bottom of the dish. Allow the plates to cool for about ten minutes, or until the media is firm and rigid. Invert the plates, if they have glass covers, and put in the 37 degree incubator. After twenty-four hours incubation, the colonies on the plates should be counted and the result recorded. Counting should be done with the aid of a reading lens having a handle and which magnifies about two and one-half diameters. . DILUTIONS. If the number of bacteria in 1 c.c. of water, exceed 250, a dilution of the sample should be made. This is done in the following manner: (a) (b) (a); (b), (c) Devolopment of Bacteria on Culture Plates. (c) (Cut furnished by New York Continental Jewell Co.) 1 c.c. of the sample is carefully measured into a bottle containing 99 c.c. of sterile water, or if this dilution is too high, 10 c.c. may be added to 90 c.c. of sterile water. This would be a 1 to 100 and 1 to 10 dilu- 22 tion, respectively. In doing routine work, the approximate number of bacteria may be estimated and the proper dilution made if necessary. The sterilized water is obtained by measuring 101 or 92 c.c. of dis- tilled water into bottles fitted with rubber stoppers covered with muslin. These bottles and contents are then sterilized in a manner similar to sterilizing media. During the process of sterilization the stoppers must be inserted loosely and about 2 c.c. is lost from each bottle. After steril- ization is complete, the stoppers should be tightly put in place and pro- tected by tying the muslin about them. The bottles should be stored in a refrigerator. Another method of making dilutions is by taking 9 c.c. of sterile water with a 9 c.c. pipette, which is also sterile, and putting it in an empty sterile bottle, adding 1 c.c. of the water to be tested, and then plating from this mixture. This would be a 1 to 10 dilution. Other dilutions can be worked out in the same manner. Before recording the result,, the number of colonies counted must, of course, be multiplied by the dilution. When making platings of unknown or badly polluted waters, two or even more plates of different dilutions should be made, in order to have one which will not be overgrown. On the other hand, too great a dilu- tion should also be avoided. Fractions of cubic centimeters may be plated but this method is not recommended. PRECAUTIONS. In making dilutions, as in all bacteriological work, the greatest care must at all times be observed to avoid contamination. The work should be done in a room where there is no severe draft and which is free from dust. All glassware and other substances that come in contact with the water to be tested and with the sterilized water, should be sterilized. The lips of bottles and test tubes should be flamed before pouring water or media. The standards of the United States Treasury Department as noted represent, of course, the maximum amount of contamination permitted. In routine plant operation, if at any time a 10 c.c. volume inoculation, of the clear chlorinated water, shows the presence of B. coli, a thorough' investigation should be made of the chlorinating apparatus, method of taking samples, and technique used in the inoculation. The temper- ature and the time occupied in sterilizing glassware should also be checked. If the water is turbid, examine the coagulation and the filters. Keep a close watch on the amount of residual chlorine. If the average number of bacteria on agar is more than 20 per cubic centimeter, with or without B. coli occasionally present, the filter underdrains or the filter beds themselves may be infected and overgrown. This condition is usually remedied by applving a strong hvnochlorite solution to the water on the filter, filtering to waste until no residual chlorine appears in the filtrate. INTERPRETATION OF RESULTS. MATERIALS AND APPARATUS REQUIRED FOR BACTERIOLOGICAL ANALYSES. 2 lbs. Lactose Bile Peptone Broth, dehydrated. 1 lb. Nutrient Agar. 23 1 lb. Nutrient Broth. Digestive Ferments Company's brands. 1 lb. Endo's Agar. 1 balance, Harvard trip style; sensitive to one-fourth gram. Capacity 1 kilo. Complete, with set of brass metric weights. 1 incubator; inside measurement at least 9x9x12 inches; regulated for 37 degrees C; oil or electrically heated. Complete with ther- mometer. 1 hot air sterilizer, double wall ; inside measurement about 10x12x10; oil or electrically heated. 1 autoclave. (Steam pressure sterilizer.) American standard. In- side dimensions, 11 inches diameter by 24 inches deep; either gas or oil heated. ' Or, in place of the autoclave: 1 Arnold steam sterilizer; cylindrical form; inside dimensions 10x9| inches. 1 two-burner stove; oil or gasoline burner. 1 two-quart double boiler. 1 two-quart aluminum kettle. 2 enameled spoons, 12 inches long. 1 water bath; circular, diameter 6 inches, with perforated plate for test tubes. 100 test tubes. Size 150x16 mm.; without lip. 100 test tubes. Size 150x20 mm.; without lip. 6 test tube baskets, dimensions 6x5x4 inches. 4 test tube supports for large tubes. 2 test tube clamps. 2 lbs. absorbent cotton. 24 4-oz. glass stoppered bottles (ifarrow neck). 12 8-oz. glass stoppered bottles (narrow neck). 24 8-oz. bottles with rubber stoppers. 3 yds. unbleached muslin. 24 pipettes, capacity 10 c.c. (not over 10 inches long). 24 pipettes, capacity 1 c.c. (not over 10 inches long). 12 pipettes, capacity 9 c.c. (not over 10 inches long). 12 pipettes, capacity 1 c.c., graduated in one-tenths (not over 10 inches long). 2 pipette boxes, dimensions 3x3x11 inches. 2 pipette boxes, dimensions 2x2x11 inches. 36 standard Petri dishes. 2 culture dish holders for standard dishes. 1 thermometer graduated in degrees to 150 degrees C. 1 thermometer graduated in degrees to 300 degrees C. 1 reading lens with handle, magnifying power 2^ diam. 1 volumetric flask, capacity 500 c.c. 1 volumetric flask, capacity 500 c.c. 1 volumetric flask, capacity 100 c.c. 1 graduated cylinder, capacity 100 c.c. 1 graduated cylinder, capacity 50 c.c. 1 glass funnel, diameter 3 inches. 1 glass funnel, diameter 5 inches. Bacteriological equipment mav be purchased from the same firms handling chemical apparatus. See list. 24 SOURCES OE WATER SUPPLIES. Before considering the construction and operation of the various types of water purification plants, it would seem advisable at this time to discuss briefly the different sources which may be drawn upon to sup- ply water to municipalities, institutions and manufactories where a con- siderable quantity of water is consumed. It is important for an oper- ator- to be familiar with the source of a supply, as much valuable in- formation may sometimes be gained in advance and this knowledge turned to good advantage by taking extra precautions in safeguarding the water which finds its way to the consumer. This applies not only to surface waters but in a lesser degree to ground waters as well. Generally speaking, there are six principal sources of supply which may be designated as follows: 1. Streams. 2. Impounding reservoirs. 3. Lakes. 4. Wells. 5. Springs. 6. Infiltration galleries. A sanitary survey of the drainage area of a surface supply is very important. This survey should be not only thorough but complete. In it should be noted all possible foci of infection located on the watershed. STREAMS. If the water is taken from a stream the sanitary survey should include the situ dion of towns, description cf the meth . Is of sewage, garbage and night-soil disposal used. Not only the larger cities and towns but the smaller hamlets as well, and even the farmsteads immediately above the pumping station should be investigated. A frequent report on the gen- eral health conditions over the whole area is desirable. If there are an unusual number of dysentery, diarrhea, typhoid or cholera cases on the catchment area above, the conscientious operator will watch his plant more carefully than otherwise. The health conditions can always be obtained by keeping in touch with the various county and city health officers on the watershed above. The water in a stream has a ten- dency to purify itself. The rate of purification, or the distance it must flow in a given stream, cannot be stated with any degree of accuracy. It depends in part on the condition of the water. If the water is highly colored or carries considerable finely divided turbidity in suspension, it will not purify itself readily. Sunshine is nature's great sterilizing agent. If the water is high in color and finely divided turbidity, tb^ ravs of the sun cannot pierce f&r and sterilization proceeds slowly. Another important factor is the depth of the stream and rate of flow. If the stream is comparatively shallow, and the rate of flow high, the water will be turned over often in its journey, and in that way all of it will be exposed to the action of light. The air also, it should be mentioned, has beneficial effects, not only in the destruction of bacteria but in the removal of odors and the oxidation of organic matter. Heavy, rapidly settling turbidity will carry down a good percentage of tbe organisms, but the real determining factor is the number of hours of exposure to direct sunlight. 25 Bacteria and domestic contamination, however, are not the only prob- lems the plant superintendent has to deal with. In a plant where alum is used as a coagulant, it is absolutely necessary that a certain amount of residual alkalinity is present in the water, otherwise no reaction will take place between the alum solution, which is of an acid nature, and the water, and of course no coagulation. If the water is normally low in alkalinity, and a series of heavy rains fall on the upper watershed, the alkalinity may take a sudden drop and fall below the safety mark where proper coagulation will take place. The amount of alum neces- sary is usually determined by the turbidity and color in the water, and these are ordinarily at their highest mark when the alkalinity is at its lowest. This phase of plant supervision will be discussed in more detail under "Chemical Treatment." Maximum daily variation in alkalinity commonly takes place during the winter months after rains, when the ground is frozen and the water has a maximum run-off. The water then has very little chance to dissolve the mineral constituents of the soil which would add to the alkalinity. IMPOUNDING RESERVOIRS. Impounding reservoirs as a source of supply are made use of ordi- narily where the rainfall is periodic or uncertain, and where no large stream or lake is convenient. Usually a reservoir of this type is formed by building a dam across the lower end of a draw or ravine. Often a small lake is made the basis of a reservoir. The holding capacity being increased by dredging or simply increasing the height of the' banks, the supply of water is commonly supplied by dry weather branches or creeks and direct drainage from the catchment area. The chemical composition of the water in an impounding reservoir is much more uniform than that in streams. For this reason, if a method of chemical treatment has been worked out which has proven satisfactory, this scheme can be followed without making any extensive changes., ■ From a bacteriological standpoint, however, the water from impound- ing reservoirs requires constant attention. The drainage area is usually small and for that reason more easily controlled from a sanitary stand- point. The water, however, after being once contaminated has much less chance to purify itself. Bacterial reduction takes place principally through sedimentation. After the coarser, heavier turbidity has set- tled, the rate of bacterial sedimentation is very slow. The action of light effects only the upper layers of water. If the water is deep, the lower strata will not be affected at all, or at most only to a very lim- ited extent. LAKES. Next in order of importance to be considered are lakes. A large lake, comparatively speaking, is without doubt the best source of supply for a municipality. From such a source the water is most uniform in chemical composition. The temperature of the water will also continue to be more uniform throughout the year. This is a great aid to the filtration plant operator. Furthermore, the consumers will become ac- customed to the taste and other qualities of the water, and as these will 26 change but very little throughout the year, cases of dissatisfaction on that score will be rare. The observance of sanitary conditions on the drainage area of the large lakes is as important as on that of the smaller ones, and should be watched and as rigidly enforced as the laws will permit. The bacterial infection, which can not be avoided entirely, will be greatly diluted, and any danger therefrom accordingly diminished. Self-purification takes place by sedimentation and by the action of the sunlight and air. The smaller lakes may be considered as being in practically the same class as the impounding reservoirs. WELLS. Wells are the most popular as supplies for the smaller towns, villages, institutions and many manufactories. Well water seldom requires treat- ment of any sort. Careful logs should be taken when the wells are drilled. A study of the logs and the geological formations in the neigh- borhood of the wells, would show whether or not there was danger of surface seepage entering the water-bearing strata that furnish the water for the wells, it may also give information as to the quantity available. SPRINGS. From time immemorial springs have been considered as the ideal source of supply. The general public too often believes that if water comes from a spring and is clear and cool, it is above reproach. Few things regarding water could be more erroneous. It is a lamentable fact that a great many cases of sickness have been caused by the water from contaminated springs because of this unfortunate popular belief. Springs, more so than any of the sources previously discussed, require a careful sanitary survey of the surrounding area and study of the probable source of the water delivered. INFILTRATION GALLERIES. Infiltration galleries were constructed more than 2000 years ago. They may properly be considered a forerunner of the slow sand filter. Only a few infiltration galleries are now in use in the United States. They are seldom recommended in this era of water purification. CONSTRUCTION AND OPERATION OF MECHANICAL WATER PURIFICATION PLANTS. It is difficult to discuss the construction of water purification plants as a general subject, because at every plant will be found problems pecu- liar or applying only to it. The subject can, however, be made more specific by dividing the purification plants into groups according to their equipment and construction. (1) OPEN GRAVITY TYPE. (2) PRESSURE TYPE. MECHANICAL FILTERS Of premier importance in the construction and operation of a plant is a study of the character of the water to he treated and filtered. This is especially true of plants with filters of the mechanical type. 27 The usual reason for treating a water with chemicals and subse- quently filtering it, is to remove suspended turbidity. There are, how- ever, other reasons also, such as removal of color, odor, taste, iron and manganese. Water is also treated chemically for the removal of exces- sive hardness. The removal of turbidity and color are accomplished in about the same manner, in so far as the treatment is concerned. First of all a chemical analysis of the water should be made. If the source of the supply is a stream, these analyses should cover, if possible, seasonal variations. If the supply is from a lake or im- pounding reservoir, the water, as has been stated before, will be much more uniform in composition throughout the year. These analyses should be made before the plans of the plant have been drawn, as the design will depend more or less upon the treatment the water requires. If the water has an alkalinity of 30 or more, it is usually easily treated with coagulants, and the results usually satisfactory, provided the plant is properly constructed and operated. If the water is more nearly neutral, with an alkalinity of 25 or less, artificial alkalinity will have to be added. This is accomplished by the addition of a solution of lime or soda ash. Lime may be added in a dry form by the use of some form of dry feed apparatus. When the dry feed is used the lime is usually fed into a tank or cistern and then carried by a stream of water to the point where it is to be applied. In this connection it should be borne in mind, that, generally speaking, only the lime in solution will add to the alkalinity of the water. For this reason, if only the saturated solution is used, a considerable saving of chemicals is effected. The detailed plans of construction of filtration plants depend, as has been stated above, on local conditions, taking into consideration the type of water, presedimentation and coagulation basins required, coag- ulants best suited, money available and other important features. No set of conditions will be found the same. Every plant should be de- signed and constructed to meet the peculiar conditions of the commu- nity it is intended to serve. As the scope of this work is limited, only a few of the typical classes of water with a description of the plant arrangement best suited for that particular class, can be discussed. In order to best serve the purpose of this pamphlet, the waters may be divided into three classes. I. Hard alkaline waters having an alkalinity of 150 P. P. M. as cal- cium carbonate: total hardness of more than 150; turbidity varying from 5000 to 10; color, 0-150. II. Soft alkaline waters, having a total hardness of 150 or less; alkalinity, 0 to 150; turbidity varying from 5000 to 10; color, 0-150. III. Acid waters, having a varying acidity; turbidity, 5000-10; color, 0-150. It is assumed that all waters requiring filtration will occasionally be subjected to bacterial infection. 28 GENERAL ARRANGEMENT OF PLANT CLASS I WATER. Water with an alkalinity of 150 or more is usually easily acted upon by a coagulant, especially alum. However, if the turbiditv is high, over 300, in the water as treated, a considerable quantity of the coagulant will be required. Ah appreciable amount of aluminum will react with or be absorbed by the particles of clay or colloidal organic matter. Much depends on the nature of the turbidity. If it consists of very finely divided particles, such as infusorial earth or FullePs earth, complete clarification in the coagulation basin is extremely difficult, and if ac- complished, will consume considerable time. For a water of this type, a presedimentation basin is to be recom- mended. This basin should be baffled in such a way that the water cannot tunnel. The time required for sedimentation again depends on the fineness and specific gravity of the turbidity. In no. case should the movement be more than 20 feet per hour. If a greater velocity is at- tained, sedimentation will be delayed. In many cases where the water may be simply diverted or the cost of pumping is low, it may be practical to construct an earthen reservoir to be used as a presettling basin. Natural ground conditions may very often be taken advantage of, and the construction cost kept down in that way. If possible the basin should be so located that the water from it will flow by gravity from it through the entire plant. If the turbidity is of such a nature and the size of the presedimentation basin such that 200 or more parts per million remain at the outlet, provision should be made for a double alum dosage; in other words, the alum should be applied at two different points. This double application of alum will be found especially helpful in cold weather; the reaction be- tween the alum solution and the water does not take place as rapidly and as vigorously as it does when the water is warm. The above is also true if the turbidity is of very fine, opalescent or milky organic nature. If the sediment is of this character when the temperature of the water is low, the difficulty of satisfactorily treating the water will be cor- respondingly increased. MIXING CHAMBERS. From the sedimentation basin the water flows into the mixing cham- ber, at the head of which it is dosed with coagulant. The best form of mixing chamber is a long baffled compartment constructed of reinforced concrete. The over and under type of baffles are considered the best. This arrangement gives the water a tumbling motion which promotes a thorough agitation. The size of the chamber and the arrangement of the baffles should be such that the movement of the water at the head of the chamber should be about four feet per second. After five min- utes at this velocity the rate of flow should be reduced to approximately two feet per second. The total time in the mixing chamber should be at least fifteen minutes. The baffles may be so arranged that the rate of the water through the chamber can be regulated to correspond with the variation in the rate of treatment. It is very important that a 29 thorough mixing of the chemical and the water is effected. The mixing should take place as rapidly as possible after the addition of the coagu- lant and before much coagulation has appeared. If this is not done, the coagulating properties of the chemical will be greatly reduced. Baffle walls may be. constructed of two-inch plank, so arranged that they can be raised and lowered, and thus adjusted to permit of the Figure 1 Figure 2 Illustrating the Principle of Coagulation in Water Treatment. Fig. 2, Clear water, and Fig. 1, water in which the floc is subsiding. (Cut furnished by N. Y. C. J. Filtration Company.) variations in the rate of operations. In warm weather when chemical reactions take place rapidly, and the floc forms in a short time, it may be advisable to further reduce the rate of flow in the lower part of the mixing chamber. In no case should a practice be made of breaking up to any great extent, the floc after it has once formed, because it usually will not recoagulate in a very satisfactory manner. COAGULATION BASINS. Too much care and thought cannot be given to the construction and operation of the coagulation basins. The coagulation basin is the heart of the plant. If this part of the plant is not operating properly, great difficulty may be experienced in obtaining a clear filter effluent. 30 The basin should be built with around the end baffles. The time required for coagulation and sedimentation varies very much indeed. The character and composition of both the dissolved and suspended matter effects this part of the water purification process more than any other. If the dissolved organic matter is high the reaction will proceed slowly, the chemical solution forming a colloidal combination with it which will settle very slowly. A low temperature of the water and a \ery fine milky turbidity will aggravate the situation further. A water of this type must be clarified in the coagulation basin, it cannot be filtered successfully until it is. This kind of a water is usually treated to the best advantage by giving it a secondary dosage of alum four to six hours before filtering. In order to do this properly, it will be necessary or at least desirable to construct another mixing chamber. This chamber need not be as extensive as the first one. However, the more thoroughly the mixing is done, the less chemical is required. The proportion of the solution to be added as the secondary application de- pends on the rate of sedimentation. If possible it should be so regulated that a little of the floc remains in the water in suspension at the time the water leaves the coagulation basin. The coagulation basin of a plant built to treat a water of the type mentioned above, should be so designed that it would have a capacity of such dimensions which would permit a flow of about 20 feet per hour. The depth and other dimensions, including the baffle arrange- ment necessary to give the water this approximate velocity, will depend on local conditions, and, of course, upon the volume of water to be treated. WATER SOFTENING. The question of whether the water of a certain supplv should be soft- ened depends, first of all, upon the hardness of the water. This must be considered in connection with the other chemical constituents, as must also the several different kinds of hardness, and the causes of each. It also depends to a considerable extent upon the percentage of water used for domestic purposes and for other uses where a soft water is desirable. The chemistry of these operations, and the analyses of waters for the chemical constituents causing hardness are beyond the scope of this pamphlet. This work must be done in a laboratory where more elab- orate equipment is available. The economic advantages of a soft water have been thoroughly studied from all angles. According to Mr. George C. Whipple a water de- teriorates in value 10 cents for every part per million of total hardness it contains. In Whipple's book, "The Value of Pure Water," the author has the following table: State. City. Source of Supply. Total Hard- ness Parts per Million. Deprecia- tion per million Gals. Maine-Waterville-Messaonskee "River .. . . 15 $ 1 50 Mainp-Amrusfa-Kennebec River ......... ;... 20 2 00 Massachusetts-Cambridge-Storage Reservoir .. . 33 3 30 Now York-Albany-Hudson River . . . . 6-1 6* 40 Pennsylvania-Philadelphia-Schuylkill River .... 179 17 90 Ohio-Toledo-Maumee River . . . . 200 20 00 Ohio-Columbus-Scioto River .... 335 33 50 31 These calculations were made with the value of soap reckoned at pre-war prices. Tn a municipal plant where the water is used for various purposes two chemicals may be used for softening. These are soda ash (sodium carbonate) and lime. The lime may be either the freshly burned kind or the hydrated lime. Under the most favorable conditions water cannot be softened much below fifty parts per million because of the solubility of calcium and magnesium carbonate. This solubility depends mainly on the tem- perature and on the other constituents present in the water. Installation for Softening Water by Permutit Process. (Courtesy of Permutit Company.) The permutit process of softening may be employed in laundries and manufactories where very soft water is desirable. By this method a water of zero hardness may be obtained. The permutit installation is quite expensive, as is also the cost of operation. For that reason it is not applicable to softening water for municipalities where much is used for flushing streets and other sanitary purposes, where hard water would answer as well. It will, therefore, hardly be necessary to consider this process in connection with the other work outlined in this pamphlet. It should be stated, however, that whenever the permutit method is used in connection with a filtration plant, the clear filtered water only should be treated. Water containing turbiditv or iron and manganese in solu- tion cannot be handled in a satisfactory manner. APPLICATION OF CHEMICALS USED FOR SOFTENING. Wherever it is deemed necessary to soften a water in connection with a water purification plant, two mixing chambers should be provided. The softening solution, whether soda ash or lime is used, should be added before the coagulating solution. A thorough mixing is required. 32 A period of thirty minutes' contact is usually sufficient for the reaction to be advanced far enough so that the coagulant may be applied. A second mixing chamber is required to mix the alum or iron solution with the water. Too much emphasis cannot be placed on the necessity of the thoroughness in mixing. Where the first cost of construction is an important factor in the establishment of a purification plant, and if, furthermore, the operation of the mixing chamber and-coagulation basin is intermittent, and electric power is cheap, it is. often advisable to install mechanical mixers. These mixers do not occupy much space, and when they are used, the expensive construction of the mixing chamber may be avoided. Mechanical mixers are quite adaptable in plants where occasionally, during periods of muddy water, a secondary dosage of alum is required. The mixer can be installed at almost any point in the coagulation basin, doing away with the second mixing chamber. The mechanical mixer does not materially decrease the head as a baffled mixing chamber does. ARRANGEMENT AND OPERATION OF FILTERS. OPEN GRAVITY TYPE. The open gravity type of filters is the most popular and practical for municipalities. They may be constructed of various sizes and may be built of wooden staves or reinforced concrete. The wooden tub filters are seldom constructed larger than for a rate of 500,000 gallons per day. The concrete filters are not so limited as to capacity. They may be built large enough to filter 4,000,000 gallons or even more per day. Ordinarily speaking, the rate or capacity of a filter is determined by the surface area of the sand bed. The rated capacity of a filter is usually reckoned on the basis of a rate of 125,000,000 gallons per acre per day. They may, of course, be operated at any rate lower than the specified capacity, the lower the better. A filter may also be operated at a capacity somewhat greater than the rated one, but this is not advisable. It must be remembered, and should be emphasized, that a filter is not merely a tub or box containing gravel and sand through which water is filtered. The bed of a filter is in reality the result or appli- cation of principles of chemistry and physics. It is the experience of every operator that a filter "works best" after it has been in service for a few hours after washing. It is a commonly known fact that a filter removes bacteria and even submicroscopic particles or atoms much smaller than bacteria and very many times smaller than the interstices or open- ings between the sand grains. Ellms, in his book on "Water Purifica- tion," makes the following statement: "A close examination of a well 'ripened' sand filter reveals the sand grains covered with a very thin film of a gelatinous character, which is most abundant near the surface of the bed, but appears to permeate it to some depth. This material is evidently in part composed of organic matter, both living and dead. It is in the colloidal form and seems to fill partially or wholly the interstices of the bed, and to be supported by the sand grains. Not all of this material between the sand particles is of an organic character. Some of it is inorganic, but it, too, is ap- parently in the colloidal form. "In a rapid sand filter, which is frequently disturbed by washing, the 33 "Modified Jewell" Gravity Filter. One of the Earlier Types. (Courtesy of N. Y. C. -J. Filtration Company.) 34 colloidal matter has less chance to develop and less opportunity for lodgment in the openings between the sand grains than in a slow sand filter. Nevertheless, when organic matter is plentiful in the water being treated, the sand grains become coated and not even violent washing of the bed will entirely remove it. Moreover, in this class of filters col- loids produced by the reaction of certain chemical compounds which are added to the water, are employed to assist the natural colloids in their straining action. "Continental" Air Wash Gravity Filter. A Later Model. "The filtering action of a sand bed must therefore be attributed to the properties of the colloids deposited or formed therein. A consideration of some of the properties of this form of matter will perhaps explain the ability of filters to perform the work which is so characteristic of them. The colloidal state of matter can be considered as a mesh struc- ture, or a sort of porous framework. Through it water diffuses readily, but suspended colloids contained in the water are held back. Thus, the bacteria, which are of a colloidal nature, are unable to pass through the mesh-like structure of the colloids attached to the sand grains, or are held back by adhering to the surface of the colloidal matter upon the sand and in the passages between the sand particles. (Courtesy of N. Y. C. J. Filtration Company.) 35 View in the St. Louis, Mo., Plant. Capacity 160,000,000 Gallons Per Day. (.Courtesy of Builders Iron Foundry.) 36 View of Operating Floor, Minneapolis, Minn., Plant. Capacity 96,000,000 Gallons Per Day. (Courtesy of Builders Iron Foundry.) 37 "The adhesion of colloidal particles to each other may be and probably is more than a mechanical attachment: that is, it may be an electrical attraction." The filters in the larger plants are usually arranged on either side of a pipe gallery, in which are the influent, wash water, sewer, filtered waste pipes, effluent, gates and connections. The strainer system in practically all of the small plants, and in many of the larger ones, con- sists of a cast iron header pipe extending across the bottom of the tub, or running lengthwise over the bottom of the rectangular tank, as the case may be. Smaller lateral pipes extend from the header-pipe to the sides of the filter. Into these laterals as well as into the header itself, the filter nozzles are screwed. The bottom of the filter is then filled in with concrete to such a depth as will cover the pipe manifold, leaving Continental Strainer System with Header and Manifolds. Strainer Nozzles in Place. (Courtesy of N. Y. C. J. Filtration Company.') only the filter nozzles exposed. If the plant is supplied with an "aiir wash" arrangement, the filter nozzles can be used to distribute the air, or a separate manifold may be put in to be used exclusively for the air. The header joins by means of the proper pipe connections to the effluent, wash water and sewer mains, and in most cases where air is used, to the air line. The filter nozzles or strainers have small per- forations through which the water must pass, and if used for air-wash- ing, they also distribute the air. Instead of filter nozzles, most of the larger plants have strainer plates arranged over valleys between blocks of concrete. They are of varying 38 lengths and of different widths, depending on the construction of the filter bottom. The plates are made of either brass, bronze or monel metal. Directly above and surrounding the filter nozzles or strainer plates is the gravel. The gravel is usually placed in five layers of different grade. The thickness of each layer is from two to three inches, and the size of the gravel in the different grades varies from one inch to one and one- half inches for the lower or largest grade, to one-fourth to one-sixteenth inch for the top or smallest grade. On top of the gravel should be placed about twenty-eight to thirty inches of sand. This sand in order to give the best results should conform to certain specifications. The writer has found that sand with the following specifications gave satis- factory results: Effective size, 0.35 mm. Uniformity coefficient, 1.60. According to Mr. Allen Hazen in his book "The Filtration of Public Water Supplies/' effective size and uniformity coefficient are defined as follows: "Effective size is such a size than which 10 per cent is smaller and 00 per cent larger than this size. "The ratio of the size of the grains, such that 60 per cent of the sand is finer than this size to the effective size, is termed the uniformity coefficient." Sand with rounded edges is to be preferred to sand with sharp edges, as the former will not pack as ouickly. WASH WATER TROUGH. The capacity of the wash water trough must be considered in con- nection with the area of the filter bed and the velocity of washing neces- sary. The height of the trough above the sand depends on the fine- ness of the sand and the rate of washing. The average height is 24 inches. LOCATION OF FILTERS. The filters should be so located with reference to the coagulation basin that the water flows by gravity from the basin onto the filters. The operating level of the filters should be the same as that of the coagu- lation basin and the influent pipes should be large enough to permit the water to move slowly on its way to the filters, and thus prevent the t!oc from being broken up. The ideal arrangement is to build the filters immediately adjacent to the coagulation basin. CLEANING OR BACK WASHING FILTERS. The earlier type of mechanical filters had agitating devices of vari- ous design. These stirring apparatuses were used in connection with the slow wash process. This method of cleaning filters has become -obsolete and is very seldom used, having given way to the high velocity system of washing. " Air may or may not be used with the high velocity wash. In plants where the' turbidity in the water treated is due more or less to a sticky mud, which in part is carried on the filters with the floc, the air wash 39 will be of great assistance in breaking up the surface layer of sticky sand and mud preparatory to turning on the water. In this case the use of air would effect a considerable saving of wash water. The use of air, however, requires a distribution system. When air is employed it is used at the rate of approximately five cubic feet per square foot per minute under a pressure of three to five pounds at the nozzle. The air application should precede the water washing. It is best to keep the sand bed covered with an inch or two of water during the air agitation process which usually occupies about four minutes. In filters where the air has a separate system of nozzles, the air and water may both be on at the same time. In this case both the volume of air and the pres- sure must be decreased, otherwise there would be danger of washing sand over into the gutter. Filter Operating Table. Minneapolis Filtration Plant. (Courtesy of Builders Iron Foundry.) The water used to wash a filter is, of course, filtered water which is applied to the filter under pressure. This pressure may be supplied from a wash water tank, erected for that purpose, back pressure from the pumps or the regular pressure from the waterworks system, depend- ing on local conditions. For the most economic washing of filters the wash water should be used at as great a velocity as possible without washing sand into the troughs or gutters. The rate will be determined to a great extent by the thickness of the gravel layer, the fineness of the sand and the height of the troughs above the surface of the sand. Ordi- narily the volume of water used under average conditions is from 12 to 15 gallons per square foot per minute, or a vertical rise in the filter of from 19 to 24 inches per minute. If it is found that this rate of washing carries sand over into the gutters the rate should be decreased. The sand lost during the washing will consist of the finer grades, so that if this loss of sand continues at each washing the effective size of the sand in the filter will be increased and the uniformity coefficient 40 lowered. In other words, the sand as a whole will become coarser and the grains more of an equal size, the filtering efficiency of the sand being correspondingly reduced. The washing should continue until the wash water runs quite dear. The quantity of wash water used at a plant depends a good deal upon the nature of the collected sediment to be removed from the sand bed, the care the operator uses and the equip- ment he has to work with. It is the writer's experience that in no case should it be necessary to use more than 4 per cent of the water filtered for washing. After the washing process has been completed, the wash water turned off and the sewer closed, the influent gate should be opened and the filter filled to its operating level. This will give the sand time to settle and the bed to become firm. If the filter has been thoroughly washed and the water from the coagulation basin does not contain much finely divided sediment, it will not be necessary to use the filtered waste. The filter effluent gate may be opened instead and the water run into the clear water well. LOSS OF HEAD GAUGES. Loss of head in a filter may be defined as the column of water neces- sary to force water through the layers of sand and gravel and through the small holes in the strainers or filter plates. As the filter continues in use after washing, the sand will become packed closer and closer, and the interstices between the sand grains, especially near the upper surface, will become filled in more and more with colloidal material and sediment filtered from the water. This condition of the filter will necessitate a greater pressure to force water through, and we say that the loss of head has increased. As the -loss of head continues to in- crease, a time will come when the filter will be unable to filter water at the rate required. It will then be necessary to wash the filter. Various types of apparatus have been designed, that when properly connected, will at all times indicate the loss of head under which a f Iter operates. All loss of head gauges are based on the principle stated above. They should be installed in every plant, as they will enable the operator to save wash water and chemicals and also keep the filters in good condition. The rate of filtration of a filter may be manually controlled by throt- tling the effluent valve or by the use of one of the many automatic rate controllers. It is practically impossible to regulate the rate accurately by the former method, so for that reason a rate controller of the auto- matic type is advised. A semi-automatic rate controller is described by Ellms in "Water Purification." This rate controller consists of a disk inserted in the discharge outlet of the filtered water pipe. This disk has an orifice of such a size that the rate of flow desired will be at the maximum rated capacity of the filter when the sand is clean and the head of water above the sand at some minimum level. If another orifice plate under a constant head of water is placed in the influent pipe of the filter, the delivery of water to the latter can be maintained at the same rate as the discharge. As the filter becomes clogged the height of the water above the sand will increase, thereby increasing the head on the outlet orifice. The gradually rising level of the water above RATE CONTROLLERS. 41 the sand becomes a direct measure of the frictional resistance produced by the clogging of the sand. When the water above the sand has reached its maximum level, the filter must, of course, be stopped and washed, if the rate of flow is to be maintained. Other types of controllers are constructed in such a way that they utilize the difference of pressure of water as it passes from the body through the throat of a Venturi tube. Still others are operated by a float arrangement which, rising and falling, operates valves that con- trol the flow of water through the apparatus. Rate controllers may at any .time be calibrated or checked in the fol- lowing manner: With the filter in operation, close the influent valve and then note the time it takes for the water on the filter to drop a definite distance. This operation should be repeated at least once. Knowing the area of-the filter, and the time consumed to lower the water a certain distance in the filter, the. volume of water filtered dur- ing the certain period can be calculated, and from this data the rate of the filter determined. PRESSURE TYPE FILTERS The pressure type of rapid sand filters are used to filter all classes of water that can be filtered by the open rapid sand filters. They are least adaptable to waters which are difficult to clarify in the coagulation basins, where some of the fine sediment remains uncoagulated and is car- ried onto the filters. Preparation of the water previous to filtration, in the modern type of pressure filters, is the same as for the open type of gravity filters. The different steps and processes need not be repeated. The pressure installation is quite well suited for manufactories and large office buildings which have their own water systems. The bac- terial efficiency is not as high as in that of the open gravity type. There is wide difference of opinion bv leading authorities as to the relative cost of operation. Mr. Geo. A. Johnson, in his pamphlet, "Rapid Sand Filtration," published in 1917, writes as follows: "Pressure filters of the rapid sand type are now operating on munici- pal supplies in this country in 140 places with a present total population estimated at 1,946,000. The combined capacity of these 140 plants is 257,200,000 gallons daily. The individual plants range in size from 100,000 gallons daily up to the 21,000,000 gallon plant at Atlanta, Ga. "Pressure filter plants, therefore, constitute 20.5 per cent of the total number of municipal rapid sand filter plants in the United States, 10.8 per ceijt of the total filtering capacity, and serve 10.6 per cent of the total population served. "The first rapid sand filters installed in this country for the purifica- tion of municipal water supplies were of the pressure type, and in the growth of the filtration art such shortcomings as these earliest filters possessed have been handed down and accepted by some as inherent defects in the process. The first pressure filter plants were without adequate facilities for proportional chemical application, and rate con- trol was a matter to be governed by the demand for water. The same history can be recorded for the gravity filter but the latter is capable of greater elasticity in individual design, and consequently it was favored in the development of new ideas. 42 "The pressure filter is particularly adapted to water problems where double pumping is an important item of expense, since with this filter one pumping may be avoided. Along the general line of filter-operating economy it is significant that the pressure filter is looked upon with con- siderable favor by private interests.' Of all the pressure filter plants operating on municipal supplies in this country, over one-third are owned by private companies. Such companies certainly operate their proper- ties as economically as possible, the first thought of the business man naturally being to furnish satisfactory service at the lowest possible cost. Such operators of pressure filter plants evidently are able to secure good service for less money than would be possible with gravity filters." The chemicals used in a water purification plant are alum (sulphate of aluminum), lime, either caustic or hydrated, iron sulphate, and soda ash or sodium carbonate. Sulphate of copper is at times used to de- stroy algae in reservoirs and impounding basins. Bleaching powder or calcium hypochlorite and liquified chlorine are employed as sterilizing agents. i All chemicals should be stored in a dry place away from the air. The iron and aluminum sulphates have a tendency to cake and form lumps. This can be avoided to some extent by keeping the chemicals dry. Quick lime must also be kept away from the air and moisture, as it will grad- ually be converted into the carbonate form. The carbonate of lime is of practically no value in the purification of water. Hydrated lime is easier to handle and store as it will not undergo changes very readily. It is, however, more expensive as about a third of it is water. Alum, iron sulphate, soda ash and quick lime may be handled in bulk. The smaller plants, as a rule, however, prefer to have their chemicals shipped in barrels. Hydrated lime may conveniently be transported and stored in paper bags. Calcium hypochlorite may be obtained in sheet iron drums varying in size from a few pounds to 600 pounds. These should also be kept in a dry place, as moist hypochlorite becomes quite corrosive. Liquefied chlorne usually comes in steel drums holding about 100 pounds of the gas. STORING AND HANDLING OE CHEMICALS. APPARATUS POP DISSOLVING AND MIXING CHEMICALS. Soda ash, alum and iron sulphate dissolve comparatively easily. The usual arrangement is to have one dissolving tank and two solution stor- age tanks. The tanks, ordinarily are made of concrete, although wooden tanks may be used. Aluminum and iron sulphate have a corrosive action on metals, so for that reason metal tubs should not be used for those two chemicals. Ii. making the solutions the common procedure is to weigh out a cer- tain amount of the chemical in the dissolving tank and run in the water in the bottom of this tank and drawing off the saturated solution near the top. In this way a solution of the chemical is effected in the least possible length of time. The solution tanks are used alternately. While one is filling, the solution is drawn from the other. Solutions of this nature have a tendency to stratify; that is, the solution on standing 43 Venturi Meter. Simple Type. (Cut furnished by Builders Iron Foundry.) 44 Venturi Meter. Type "M" Register-Indicator-Recorder. (Courtesy of Builders Iron Foundry.) 45 will become much more concentrated or stronger near the bottom. It is therefore necessary that some form of stirring apparatus be installed in connection with each tank, otherwise a uniform application of the chemicals cannot be made. In slaking quick-lime the slaking tank should be equipped with a stirring device. Hot water is also an aid to more rapid action. When using hydrated lime the slaking tank may, of course, be dispensed with. Lime may be applied as lime water or as milk of lime. When lime water is used a comparatively large tank storage capacity is required. When the lime is added as milk of lime it i>s usually fed into a dissolving tank or basin into which a constantly flow- ing stream of water is directed. This water dissolves the lime and car- ries it to the point of application. All tanks and basins used for the dissolving and storage of chemical solutions should be elevated, so that the solution will flow by gravity to all points desired. It is much more practical to pump the water used; in dissolving the chemicals than to pump the chemical solution. The proper application of the chemical solutions in a water purifica- tion plant is one of the most important steps in the entire process and requires apparatus which can be adjusted and regulated so as to deliver- definite quantities of solution. For small plants, where the rate of treatment continues constant for periods of time, one of the many APPLICATION OF CHEMICAL SOLUTIONS TO THE WATER. Chemical Solution Feeding Device. Constant Level Orifice Box. forms of constant level orifice feed tanks may be used. In plants where- the volume of water treated is controlled by reciprocating pumps tliat vary considerable in rate of pumpage, proportional solution feed pumps may be attached to the pumps in such a way that the volume of solution used will vary directly with the rate of pumpage. In the large plants the chemical solution feed is nearly always controlled by a Venturi tube connected to a master-controller. A change in chemical treatment, that is, in terms of grains per gallon, may be effected by either increasing the strength of the solution or by increasing the volume of solution applied. All piping used in the transfer of chemical solutions should be of ample size, and installed so as to carry the solution along the most (Courtesy of N. Y. C. J. Filtration Company.) Chemical Solution Regulating Valve. (Courtesy of N. Y. C. J. Filtration Company.) Proportional Chemical Solution heeding Device. View showing Controlling Apparatus. • (.Courtesy of N. Y. C. J. Filtration Company.) 47 Proportional Chemical Solution Feeding Device. View Showing Float Tubes. (Courtesy of N. Y. C. J. Filtration Company.) direct route. Iron and aluminum sulphate have a corrosive action. Rubber steam hose has been used successfully for these solutions. Cast iron pipe painted with graphite paint, and open cast iron troughs may very often be used to advantage for all solutions. CHEMICAL DEY FEED APPARATUS. During the last few years quite a number of dry feed apparatus have been devised. When using the dry feed apparatus it is necessary that the chemicals are pulverized into a fine powder before being fed to the machine. The usual principle on which they are constructed is the worm gear feed. The speed of this o-ear can be regulated and in that way the amount of chemical controlled. This type of feed is especially adapted to feeding hydrated lime. The powdered chemical is fed from a hopper shaped bin over the worm. It is well to have an agitator near the bottom of the hopper to keep the chemical in contact with the screw. All finely pulverized chemicals have a tendency to choke small passages or orifices. This form of feed is not suitable for OUTLET HOPPER FEEDING CHAMBER Dry Chemical Feed Machine. (.Courtesy of N. Y. 0. J. Filtration Company.) 49 soda ash, as that chemical has a tendency to cake. When this type of apparatus is used the chemical should not be fed into the mixing cham- ber direct, but should be run into a tank or basin where there is con- siderable agitation which would prevent any sedimentation on the bot- tom and would promote solution. From the solution tank a stream of water should carry the dissolved chemical to the head of the mixing chamber or to the mechanical mixer. The fact should be borne in mind that only the dissolved chemicals will be effective for the purpose for which they are applied. CLASS II WATERS. This class of water was defined as including waters having an alka- linity of 0 to 150; total hardness of 150 or less; turbidity varying from 5000 to 10; color, 0 to 150. The open gravity type of water purification plant, arrangement and equipment of which were described on the foregoing pages as being suitable for treating waters of Class I, can be used satisfactorily for Class II also. When treating a water, the alkalinity of which may decrease more or less suddenly until 0 alkalinity is reached, great care must be exercised not to overtreat with the coagulant, or an acid or neutral condition of the water will be established and no coagulation can take place. When water is quite clear and does not contain much dissolved or colloidal or- ganic matter, each grain per gallon of alum added in the treatment pro- cess will decrease the alkalinity approximately eight parts per million. Iron sulfate will decrease it six parts per million. In general practice, with waters that usually have an alkalinity of 30 or less, and that may decrease in alkalinity suddenly to a point of neutrality, an attempt should be made to keep the residual alkalinity in the filtered water at least as high as ten parts per million. If the water has an alkalinity of 25, carries considerable turbidity, or is high in color and requires three or more grains per gallon of alum, artificial alkalinity will have to be added to the water before applying the alum. Either soda ash, quicklime or hydrated lime may be used for this purpose. In practice the. alkalinity is raised about twelve parts per million for every grain per gallon of lime, in the form of oxide, added. If hydrated lime is used about 40 per cent more must be added to obtain the same results. Tests for alkali nitv of the raw, the lime treated water and the filter effluent should be made frequently, so that the chemicals will not be used in greater quantity than is necessary, thus keeping down the expense of treatment. In so far as removal of color is con- cerned the same processes and manipulations as are followed in the removal of turbidity mav be pursued. Waters with an alkalinity of 35 to 150, which have no unusual con- stituents, or industrial contamination, are most easily handled, as 'only some form of coagulating treatment is necessary and accidental over- dosing is almost impossible. When sulphate of iron is used as a coagulant, this treatment is al- ways preceded by a preliminary dosing with lime. One grain of quick- lime or three-fourths of a grain of hydrated lime to effect proper reaction. 50 CLASS III WATERS. The waters of this class are more or less acid in character, and of varying turbidity. They usually contain considerable carbon dioxide. Municipalities are rarely compelled to use this type of water as a source of supply. The fact that an acid water dissolves the mineral constit- uents of rocks and those found in the soil gives these waters a high mineral content. This solution process also tends to make the water neutral. The same plant arrangement is applicable as in the case of hard alkaline waters where softening is practiced. The application of chemicals may be carried out in two ways: 1. The coagulant may be added to the acid water and then precipi- tated with lime or soda ash, or 2. Artificial alkalinity may be applied first, followed by the co- agulant. Iron sulphate can seldom be used as an acid water, quite often con- tains a good deal of organic matter, and iron sulphate will, in that case, impart color to the water. The addition of alum to the acid water, fol- lowed by lime or soda ash solution gives a good floc. The usual method, however, is to treat the water with an alkali, lime or soda ash, in sufficient quantities to create artificial alkalinity, and then dose with alum, making frequent alkalinity tests and regulating the chemicals in such a way that, after securing proper coagulation and sedimentation, some residual alkalinity is present in the filtered water. IRON AND MANGANESE REMOVAL. Many well waters and springs are unfit for domestic purposes be- cause of the presence in them of iron and manganese, or both. The water when first drawn from the well mav be perfectly clear, while after being exposed to the air in a bucket or other receptacle for even a short time, the water becomes turbid and a red or brownish deposit forms on the sides and bottom of the containing vessel. The turbidity formed consists principally of the oxides of iron or manganese, or pos- sibly a mixture of both. The presence of iron and manganese in a water is due ordinarily to the high free carbonic acid content of the water and the fact that the water must have passed through earth or rock strata containing the metals, and holding them in solution. Dissolved organic matter may also be present and aid in keeping them in a dissolved state, but usually this is only incidental. It has been found that the removal of the free carbon dioxide pre- cipitates the iron and usually the manganese. This fact has been taken advantage of for the practical removal of these metals. The carbonic acid may be removed according to two different methods of procedure: 1. By aeration of the water. 2. By treating the water with lime or soda ash. Aeration may be accomplished by subjecting the water to the action of air. by passing it over a series of cascades, forcing it out over foun- tains and then passing through trickling filters made up of coke or other porous material. The action of the air will remove the free car- bon dioxide. After aeration the water is run into a sedimentation basin 51 from which it is passed through sand filters. A period of about six hours should be allowed in the sedimentation basin, as the precipitation of manganese takes place more slowly than that of the iron. The removal of the carbon dioxide may also be effected by treating the water with small amounts of soda ash or lime in the usual manner, allowing about six to ten hours for sedimentation and then filtering through sand filters. The methods indicated above are for waters where only the removal of iron and manganese is desired, the water being satisfactory in other respects. The presence of these metals is a great detriment to water intended for domestic uses especially when used in textile industries and laundries. Both iron and manganese are usually easily removed. Much dissolved organic matter will delay precipitation and sedimentation and the water will require more vigorous treatment. When the manganese is in the form of sulphate and the iron is comparatively low, aeration and filtra- tion will not remove all of the manganese. It will then be necessary to use lime after the water has been aerated. Allow six to ten hours in a sedimentation basin, and then filter through a sand filter. This pro- cedure will usually prove satisfactory and remove the nuisance. STERILIZATION OF WATER. We have on the preceding pages considered briefly some of the most important features of a water purification plant of the open gravity type, and have hastily reviewed in a very general way some of the other processes involved in water purification, with especial reference to a hard alkaline water carrying varying amounts of turbidity. Sterilization or disinfection of water is the simplest and at the same time the most important feature or process in connectipn with a water purification plant. The principle is the same no matter what the type of plant or character of the water mav be. The two chief chemicals used at present are calcium hypochlorite and liquefied chlorine gas. CALCIUM HYPOCHLOBITE (Ca(OCl)2). Calcium hypochlorite or bleaching powder was the first sterilizing agent used for the disinfection of water on a commercial scale. The earliest practical application we have any record of was that used on the public water supply at Maidstone, England, in 1897, following a typhoid epidemic. This chemical so well known at this time is prepared by passing chlorine gas over moist air-slaked lime. In water disinfecting practice definite quantities of the hypochlorite are mixed with water in a mixing tank and from there run in a stock solution tank. The super- natant solution is applied to the water in the same manner as the alum or iron sulphate solution, being measured .and regulated by a constant level orifice box or other control device. The clear hypochlorite solution is nearly always added to the filtered water. When treating a water which is high in dissolved organic matter it sometimes becomes advis- able to chlorinate the water just as it enters upon the filters. By so doing the slight taste that sometimes develops in a water when chlorine and dissolved organic matter combine, may usually be avoided. 52 Liquefied Chlorine Installation for Sterilizing Water. Manual Control, Solution Feed. (Courtesy of Wallace-Tiernan Company.) 53 feedWater Gauge Chlorine Check Wive Solution Jar Head Jet Orifice Cleaner Feed Water Supply Line FeedWaterValve Water Pressure Reducing Valve Strainer Globe Valve Water Seal Water Seal Waste Pulsating Meter Solution Outlet Tube Solution Jar Solution Receiver Solution Line Tank Valve Chlorine Tank Tank Pressure Gauge Chlorine Control Vblve Pressure Compensator •{Compensator Cap FlexibleTank. Connection. Auxiliarylank Valve Solution Valve Suction. Main Manual Control, Solution Feed, Chlorinating Apparatus Showing Detailed Construction. (Courtesy of Builders Iron Foundry.) 54 LIQUEFIED CHLORINE GAS. The cumbersome though effective method of disinfection with calcium hypochlorite has gradually given wav to the equally effective but simple process of disinfection with liquefied chlorine gas. This gas is made from salt by the electrolytic decomposition method. The gas is com- pressed into steel drums usually holding about 100 pounds. Several types of apparatus, of all sizes, suitable for any plant condition, have been designed and are on the market. These chlorinating machines permit of almost perfect control. The gas, like in the case of calcium hypochlorite, is usually applied to the clear filtered water, if possible in the suction line to a pump, or any place where it would be thoroughly mixed with the water. The • chemistry of water disinfection where chlorine gas is used is somewhat complicated. The final step being the liberation of nascent oxygen to which is due the destruction by oxidation of microorganisms such as bacteria. Very recently another method has been tried out. This consists of applying a mixture of calcium hypochlorite and ammonia to the water. It is known as the chloramine method. Up to this time, however, it is still in somewhat of an experimental stage. ULTRA VIOLET RAYS. Still another method which might be mentioned is the ultra violet ray process. The theory of. disinfection of water by this method is based on the well known fact that certain of the rays of sunlight have a decided bactericidal action. These rays are known as the ultra violet rays, and are present in artificial light as well as in sunlight. In the method of disinfection as applied to water, quartz lamp globes are used in place of glass globes. Light is furnished by a mercury vapor arc within the quartz globe. Quartz is used in place of glass in the manu- facture of the lamps because it will permit the passage of the ultra violet rays while ordinary glass will not. When used in a practical way, the lamps are arranged in a baffled chamber, in such a way that all water will have to pass near the lamps and thus become exposed to the direct ra^s. The final sterilizing effect is the same as that produced bv chlorine gas: that is, the liberation of nascent oxygen in the water. From an aesthetic point of view this method leaves nothing to be desired. The first cost, however, is ouite high comparatively speaking, and only a perfectly clear water can be treated successfully. The cost of operation depends on the price of electric energv. BED WATER. Nearly every waterworks superintendent and filter plant operator has had some disagreeable experience with red water troubles. This may or may not have been the fault of the persons held responsible. Waters with a high chloride content have a decided solvent action on iron pipe. Some iron pipe is more soluble than other pipe of the same appearance. It has been observed that red water may be found in a certain block on a street while in the block preceding, served by the same feeder line, the water is clear. Investigators of this subject have come to the gen- eral conclusion that solvent action may be caused by galvanic currents 55 formed between metallic elements in solution in the water and others present in the pipe. Corrosive action may also be caused by excessive quantities of carbon dioxide present in the water. When this water is the effluent from a filter plant, this condition can easily be corrected by the addition of a small quantity of lime .as a part of the treatment. Chemical tests for free carbon dioxide will indicate the amount of carbonic acid contained in the water. It is ordinarily considered that five parts per million or less will cause no harm. A TYPICAL MONTHLY REPORT OF A TEXAS FILTRATION PLANT, SHOWING REDUCTION IN BACTERIAL CONTENT. Days of Month. Raw. Settled. Filtered. Chlorinated. 1 15,000 8,000 600 48 2 16,000 8,500 500 14 3 10,000 5,000 270 9 4 2,000 1,000 250 11 5 20,000 15,000 400 22 6 2,900 1,100 300 4 7 3,800 1,700 80 9 8 1 ,400 900 20 5 9 1,600 200 18 11 10 2,000 175 27 21 11 2,300 150 12 4 12 1,000 200 22 5 13 800 250 20 21 14 300 165 14 8 15 300 100 11 10 16 175 90 21 8 17 250 95 9 5 18 250 100 10 6 19 300 175 7 6 20 1,000 300 90 4 21 600 200 25 5 22 250 175 27 5 23 200 75 11 5 24 150 30 2 5 25 100 50 5 4 26 75 40 5 3 27 75 35 9 4 28 125 75 14 1 29 120 60 3 2 30 120 75 12 2 31. .• 125 80 4 3 Maximum 20,000 15,000 600 48 Minimum 75 30 2 1 Average 2,687 1,432 90 9 Courtesy of Dr. W. T. Gooch, Chemist, Waco Water Works. REFERENCES CONSULTED. Standard Methods of Water Analysis-A. P. H. A., 1917. The Value of Pure Water-Whipple. The Filtration of Public Water Supplies-Allen Hazen. Rapid Sand Filtration-Geo. A. Johnson. Water Purification-Films. Underground Waters for Commercial Purposes-Rector. Water Purification Plants and Their Operation-Stein. 56 APPENDIX The capacity of a cylinder may be found by multiplying the diam- eter by itself (squaring the diameter), then multiply by the length and by 5.875 when the dimensions are in feet, or by 0.0034 when in inches. The result will be in gallons. To find the volume of tanks or cisterns, multiply the square of the diameter, in feet, by the depth in feet, then multiply by 47 and divide by 8. The result will be the capacity in gallons. The capacity of rectangular tanks in gallons is equal to the height times the length times the breadth in feet, divided by 7.48. The weight of water in a circular tank is equal to the capacity in cubic feet multiplied by 62.25. A column of water one foot high exerts a pressure of 0.433 pounds per square inch, or 62.35 pounds per square foot. To double the diameter of a pipe means to increase its holding capacity four times. A flow of one cubic foot per second equals 448.31 gallons per minute or 646,317 gallons per 24 hours. To measure the velocity of a stream, lay off a convenient distance, say 100 feet, on the bank of the river parallel with the direction of flow. Place partially submerged floats in the current and note the time it takes to pass the 100 feet distance. This observation should be re- peated several times, with the float placed in different parts of the .stream, to get the average velocity across the river at that particular point. If a surface float is used in the middle of the stream, multiply the velocity thus obtained by 0.83 to get true average velocity. To get the volume of flow, determine the depth of the water at 8 or 10 points across the stream at equal distances between points. Add together all the depths in feet and divide by the number of observations, the result will be the average depth. Multiply this by the rate of the current in feet per minute. The result will be the volume of water passing the particular point in cubic feet per minute. The number of cubic feet multiplied by 7.48 will be the number of gallons. To find the capacity of the cylinder of a pump, multiply the area in inches by the length of the stroke in inches, which will give the total number of cubic inches, divide this, figure by 231 and the result is the -capacity in gallons. To find the horse power necessary to elevate water to a given height, multiply the total weight of column of water in pounds by the velocity in feet per minute and divide the product by 33,000. Add 25 per cent •for friction. The consumption of fuel averages 7.5 pounds of coal, or 15 pounds of dry wood, for every cubic foot of water evaporated. USEFUL INFORMATION. 57 CONVERSION TABLES. 1 grain per IT. S. gallon equals 17.1 parts per million. 1 part per million equals 0.058 grains per IT. S. gallon. 1 part per 100,000 equals 0.585 grains per U. S. gallon. 1 pound avoirdupois equals 7000 grains. 1 pound avoirdupois equals 453.6 grams. 1 gram equals 15.4 grains. 1 yard equals 0.914 meters. 1 foot equals 30.47 centimeters. 1 inch equals 2.53 centimeters. 1 inch equals 25.3 millimeters. 1 gallon equals 231 cubic inches. 1 gallon equals 3785 c.c. (cubic centimeters). 1 litre equals 1.056 quarts. 1 quart equals 946 c.c. 1 pint equals 473 c.c. 1 ounce equals 29.67 c.c. 1 cubic foot of water weighs 62.32 lbs. at 39 degrees F. 1 cubic foot of water equals 7.5 gallons. 1 gallon water weighs 8.33 pounds. 1 ton equals 267.38 gallons of water. 1 ton equals 35.74 cubic feet of water. 1 atmosphere equals 14.7 pounds per square inch. 1 atmosphere equals 2116.3 pounds per square foot. 1 atmosphere equals 33.947 feet of water at 62 degrees F. 1 atmosphere equals 30.0 inches of mercury at 62 degrees F. Degrees Centigrade equal degrees Fahrenheit minus 32 divided by 9/5. TABLE I. HEAD AND PRESSURE EQUIVALENTS. Head of Water in Feet and Equivalent Pressure in Pounds Per Square Inch. Feet Head. Pounds Per Sq. In. Feet Head. Pounds Per Sq. In. Feet Head. Pounds Per Sq. In. 1 0.43 55 23.82 • 190 82.29 2 . 0.87 60 25.99 200 86.62 3 1.30 65 28.15 225 97.45 4 1.73 70 30.32 250 108.27 5 2.17 75 32.48 275 119.10 6 2.60 80 34.65 300 129.93 7 3.03 85 36.81 325 140.75 8. . 3.40 90 38.98 350 151.58 9. . . 3.90 95 41.14 375 162.41 10. . . 4.33 100 43.31 400 173.24 15. . . 6.50 110 47.64 500 216.55 20 • 8.66 120 51.97 600 259.85 25... . 10.83 130 56.30 700 303.16 30 12.99 140 60.63 800 346.47 35 15.16 150 64.96 900 389.78 40 17.32 160 69.29 1,000 433.09 45 19.49 170 73.63 50 21.65 180 77.96 58 Pounds Per Sq. In. Feet Head. Pounds Per Sq. In. Feet Head. Pounds Per Sq. In. Feet Head. 1 2.31 55 126.99 180 415 61 2 4.62 60 138.54 190 438.90 3 6.93 65 150.08 200 461.78 4 9.24 70 161.63 225 519.51 5 11.54 75 173.17 250 577.24 6 13.85 80 184.72 275 643.03 7 16 16 85 8 18.47 90 207'81 325 by 2. by 750 41 9 20.78 95 219.35 350 808.13 10 23.09 100 230 90 375 RAR RQ 15 34.63 110 253.98 400 922.58 20 46. 18 120 277 07 500 1 1 Rd. zlR 25 57.72 125 288.62 30 69.27 130 300.16 35 80.81 140 323.25 40 92.36 150 346.34 45 103.90 160 369 43 50 115.45 170 392.52 Pressure in Pounds Per Square Inch and Equivalent Head of Water in Feet. TABLE II. DISCHARGE EQUIVALENTS. Gallons Per Minute. Cubic Feet Per Minute. Gallons Per Hour. Gallons Per 24 Hours. Barrels Per Minute, 42 Gal. BB1. 10 1.3368 600 14.400 0.24 12 1.6042 720 17,280 0.29 15 2.0052 900 21,600 0.36 18 2.4063 1,080 25.920 0.43 20 2.6733 1,200 28,800 0.48 25 3.342 1,500 36,000 0.59 27 3.609 1,620 38,880 0.64 30 4.001 1,800 43,200 0.71 35 4.678 2,100 50,400 0.83 36 4.812 2,160 51,840 0.86 40 5.348 2,400 57,600 0.95 45 6.051 2,700 64,800 1.07 50 6.684 3,000 72,000 1.19 60 8.021 3,600 86,400 1.43 70 9.357 4,200 100,800 1.66 75 10.026 4,500 108,000 1.78 80 10.694 4,800 115,200 1.90 90 12,031 5,400 129,600 2.14 100 13.368 6,000 144.000 2.39 125 16.710 7,500 180,000 2,98 135 18.046 8,100 194,400 3.21 150 20.052 9,000 216,000 3,57 175 23.394 10,500 252,000 4.16 180 24.062 10,800 259,200 4.28 200 26.736 12,000 288,000 4.76 225 30.079 13,500 324,000 5.35 250 33.421 15,000 360,000 5.95 270 36.093 16,200 388,800 6.43 300 40.104 18,000 432,000 7.14 315 42.109 18,900 453,600 7.5 360 48.125 21,600 518,400 8.57 400 53.473 24,000 576,000 9.52 450 60.158 27,000 648,000 10.7 500 66.842 30,000 720,000 11.9 540 72.186 32,400 777,600 12.8 600 80.208 36,000 864,000 14.3 630 84.218 37,800 907,200 15.0 675 90.234 40,500 972,000 16.0 720 96.25 43,200 1,036,800 17.0 800 106.94 48,000 1,152,000 19.05 900 120.31 54,000 1,296,000 21.43 1,000 133.68 60,000 1,440,000 23.8 59 Given Flow in Gallons Per Minute Over a Weir 12 Inches Wide. TABLE III. WEIR TABLE. Depth Gallons Depth Gallons Depth Gallons in Inches. Per Min. in Inches. Per Min. in Inches. Per Min. 1 36 4J4 375 8 Ji 900 1 Ji 50 5 405 8 Ji 939 1 Ji •' 66 5Ji 436 9 979 1 Ji 84 5 Ji 468 9 Ji 1,020 2 . . 102 5 Ji 500 9 Ji 1,062 2 Ji 122 6 533 9 Ji 1,104 2 Ji 143 6 Ji 567 10 1,147 2 Ji 165 6 Ji 601 10Ji 1,190 3 188 6 Ji 636 10Ji 1,234 3 Ji 212 7 672 10 Ji 1.279 3 Ji 237 7 Ji 708 11 1,323 3 Ji 263 7Ji..: 745 11 Ji 1,369 4 290 7 Ji 783 11 Ji 1,414 4 Ji 317 8 821 11 Ji 1,461 4 Ji 346 8 Ji 860 12 1,508 TABLE IV. THEORETICAL HORSE-POWER REQUIRED TO RAISE WATER TO DIFFERENT HEIGHTS. Gals. Feet Min. 5 10 15 20 30 40 50 60 75 90 100 150 200 300 400 5 .006 .012 .019 .025 .037 .05 .06 .07 .09 .11 .12 .19 .25 .37 .50 10 .012 .025 .037 .050 .075 .10 .12 .15 .19 .22 .25 .37 .50 .75 1.00 15 .019 .037 .056 .075 . 112 . 15 .19 .22 .28 .34 .37 .56 .75 1.12 1.50 20 . 025 .050 .075 .100 .150 .20 .25 .30 .37 .45 .50 .75 1.00 1.50 2.00 25 .031 .062 .093 . 125j .187 .25 .31 .37 .47 .56 .62 .94 1.25 1.87 2.50 35 .043 .087 .131 .175 .262 .35 .44 .52 .66 .79 .87 1.31 1.75 2.62 3.50 40 . 050 . 100 .150 .200 .300 .40 .50 .60 .75 .90 1.00 1.50 2.00 3.00 4.00 45 .056 .112 .168 .225 .337 .45 .56 .67 .84 1.01 1.12 1.69 2.25 3.37 4.50 50 . 062 . 125 . 187 .250 .375 .50 .62 .75 .94 1.12 1.25 1.87 2.50 3.75 5.00 60 . 075 .150 .225 .300 .450 .60 .75 .90 1.12 1.35 1.50 2.25 3.00 4.50 6.00 75 .093 .187 .281 .375 .562 .75 .94 1 .12 1.40 1.69 1.87 2.81 3.75 5.62 7.50 90 . 112 .225 .337 .450 .675 .90 1 .12 1.35 1.68 2.02 2.25 3.37 4.50 6.75 9.00 100 . 125 .250 .375 .500 .750 1.00 1.25 1.50 1.87 2.25 2.50 3.75 5.00 7.50 10.00 125 .156 .312 .469 .625 .937 1.25 1.56 1.87 2.34 2.81 3.12 4.69 6.25 9.37 12.50 150 .187 .375 .562 .750 1.125 1.50 1 87 2.25 2.81 3.37 3.75 5.62 7.50 11.25 15.00 175 .219 .437 .656 .875 1.312 1.75 2 19 2 62 3.28 3.94 4.37 6.56 8.75 12.12 17.50 200 .250 .500 .750 1.000 1.500 2.00 2.50 3.00 3.75 4.50 5.00 7.50 10.00 15.00 20.00 250 .312 .625 .937 1.250 1.875 2.50 3 12 3.75 4.94 5.62 6.25 9.37 12.50 18.75 25.00 300 . 375 .750 1.125 1.500 2.250 3.00 3 75 4.50 5.62 6.75 7.50 11.25 15.00 22.50 30.00 350 .437 .875 1.312 1.750 2.625 3.50 4.37 5.25 6.56 7.87 8.75 13.12 17.50 26.25 35.00 400 . 500 1.000 1.500 2.000 3.000 4.00 5 00 6.0(1 7.5C 9.0C 10.00 15.00 20.0C 30.00 40.00 500 .625 1.250 1.875 2.500 3.750 5.00 6.25 7.50 9.37 11.25 12.50 18.75 25.0037.50 50.00 To allow for usual bends and friction, increase these figures at least 25 per cent. INDEX A Page- Acidity, total in water 10 Alkalinity in water 8 Phenolphthalein method 8 Methyl orange method 8 Erythrosine method 8 Apparatus for dissolving and mixing chemicals 42 Apparatus required for chemical analysis of water 12 Apparatus required for bacteriological examination of water , 15- Appendix 56 B. Coli, presumptive test for 18- Bacteria in water 14 Bacteria, total number of 19 Bicarbonate in water 9 B C Chemical dry feed apparatus 47 Chemical solutions, application of, to the water 45 Chemicals, application of, used for softening 31 Chemicals required for chemical analyses of water 13- Chemicals, storing and mixing of 42 Chloride in water 11 Coagulation basin ' 29 Color in water 5 Conversion tables ' ......................... 57 Culture Media, materials for 15 Culture Media, preparation and sterilization of 16 D Dilutions 21 Equivalents, table I, head ami pressure 57 Equivalents, table II, discharge 58. E Filters, arrangement and operation of 32" Filters, cleaning or back washing 38 Filters, location of 38 Filters, mechanical 26 Firms dealing in chemical apparatus, etc 13 Free carbon dioxide in water 10 Free mineral acids in water 11 F Glassware, sterilization of :... 15 G H Hardness, determination of total, in water 7 Soap method 7 Hardness, permanent ...... 8' Hardness, temporary 8 Hydroxide, radical (OH) 10 I Introduction 2 Iron and manganese removal 50 62 Page Loss of head gauges 40 L M Mixing chamber 28 N Normal carbonate (CO3) 9 Odor in water 6 Cold 6 Hot 6 0 P Plant, general arrangement of, for class I water 28 Rate controllers 40 Red water 54 Report, a typical monthly 55 Residual chlorine in water 12 Ortho-tolidin test for 12 Starch-iodine test for 12 Results, interpretation of 22 R Samples, collection of 3 s T Temperature, method of taking 3 Treatment required, class II waters 49 Treatment required, class III waters 50 Turbidity in water 3 Platinum wire method of determining 4 Standard turbidity method 5 Turbidity rod, method of graduating 4 u 'Useful information 56 Wash water troughs 38 Water, bacteriological examination of 15 Water, chemical examination of . . 6 Water, classification of . 27 'Class I 27 Class II 27 Class III 27 Water, physical examination of 3 Water purification plants, construction and operation of mechanical 26 Water,- softening 30 Water, sterilization of 51 Calcium hypochlorite 51 Liquefied chlorine gas 54 Ultra violet rays 54 Water supplies, sources of 24 Streams 24 Impounding reservoirs 25 Lakes 25 Wells 26 Springs 26 Infiltration galleries 26 Weir table, table III 1 „ 59 w