/ '' 
 
 y yy y y 
 y yy y y 
 
 y /yy yy y 
 
 ^ " yy y 
 
 
 
 
 
 
 
(Qatnell HmnErattg Cibratg 
 
 aittiaca, S7ew Inch 
 
 THE LIBRARY OF 
 
 EMIL KUICHLING, C. E. 
 
 ROCHESTER. NEW YORK 
 
 THE GIFT OF 
 
 SARAH L. KUICHLING 
 
 1919 
 
 ENGINEERING LIBRARY 
 
Cornell University Library 
 TD 430.N62 
 Water supply considered mainly '['O'" ^ «= 
 
 3 1924 004 975 474 ™r,. 
 
Cornell University 
 Library 
 
 The original of tiiis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924004975474 
 
WATER SUPPLY 
 
 CONSIDERED MAINLY FROM A 
 
 Chemical and Sanitary Standpoint. 
 
 BY 
 
 WM. RIPLEY NICHOLS, 
 
 PROFESSOR AT THE MASSACHUSETTS INSTITUTE OF TECHNOLOGY. 
 
 NEW YORK: 
 
 JOHN WILEY & SONS. 
 
 1883. 
 
PRESS OF J, J. LITTLE 4 CO., 
 NOS. 10 TO 20 ASTOR PLACE, ^EW 
 
PREFACE. 
 
 The following pages contain, somewhat amplified, the sub- 
 stance of a course of "Lectures on Water Supply " which the 
 author has been in the habit of delivering before certain classes 
 at the Institute of Technology. It is primarily as an aid to 
 engineering and other students at this and similar institutions 
 that the book is printed. It is hoped, however, that the book 
 will be found of service to young engineers, to persons in charge 
 of water works, to water committees, and to others who are 
 interested in the matter of water supply. 
 
 The aim is not to present a complete treatise on water supply 
 for the civil engineer, nor a treatise on water analysis for the 
 chemist, nor a treatise on mycology for the botanist, and certainly 
 not a treatise on sanitary science for the physician, but, rather, 
 to occupy a territory which encroaches on the fields of these and 
 other professions and which belongs exclusively to no one alone 
 — ground, in fact, with which all who are professionally inter- 
 ested in water supply must be more or less familiar. 
 
 The metric system of weights and measures is used, as well as 
 the English ; tables for the conversion of one system into the 
 other will be found at the end of the volume. In the nomen- 
 clature of chemical substances, the old and more famihar terms 
 are generally — although not exclusively employed — such as car- 
 bonate of soda and not sodic carbonate, sulphate of Hme rather 
 than sulphate of calcium. 
 
 The author has quoted freely from other works on the sub- 
 ject, and from his own earlier reports, now mostly out of print. 
 
IV PREFACE. 
 
 He would acknowledge especial indebtedness to the Reports of 
 the Rivers Pollution Commission, and to Fischer's chemische 
 Technologie des Wassers, and regrets that Wolffhugels Wasser- 
 versorgung did not come to hand until the manuscript was in 
 the hands of the printer. 
 
 Massachusetts Institute of Technology, 
 Boston, Mass., May, 1883. 
 
TABLE OF CONTENTS. 
 
 PAGES 
 
 Introductory Chapter. — Solution i- i6 
 
 Solution of solid substances. — Saturated and supersaturated solu- 
 tions. — Condition of mixed solutions. — Various means of hastening 
 solution. — Effects of dissolved solids. — Solution of gases. — Supersat- 
 urated solutions of gases. — Facilitating gaseous solution. — Solubility 
 of liquids in water. — Distinction between solution and suspension. 
 
 Chapter I. — Drinking Water and Disease 17-28 
 
 Chapter II. — Water Analysis 29- 47 
 
 Suspended matter. — Dissolved gases. — Total solids in solution. — 
 Chlorine. — Hardness. — Combined nitrogen. — Organic matter. — ■ 
 Standards of purity. — Popular tests. — Collection of samples. 
 
 Chapter III. — -Rain Water as a Source of Supply. 48- 55 
 
 Examination of cistern water. — Natural and artificial ice. — Chemical 
 examination of ice. 
 
 Chapter IV. — Surface Waters as Sources of Sup- 
 ply 56- 78 
 
 Turbidity of streams. —Pollution of streams. — Self-purification of 
 streams. — Oxidation. — Deposition. — Dilution. — Prevention of pollu- 
 tion. 
 
 Chapter V. — Surface Waters as Sources of Sup- 
 ply {continued^ 79-104 
 
 Animal and vegetable life. — Odors and tastes. — Temperature. — Ex- 
 amination of surface waters. — ^Variation of surface waters. 
 
 Chapter VI.— Ground Water as a Source of Sup- 
 ply 105-132 
 
 Methods of utilizing ground water. — Effects of pumping on ground 
 water. — Driven wells. — Natural filtration. — Examination of ground 
 water. — Pollution of domestic wells. — Examination of wells. 
 
VI TABLE OF CONTENTS. 
 
 PAGES 
 
 Chapter VII. — Deep Seated Water as a Source 
 OF Supply 133-14S 
 
 Artesian wells. — Deep wells. — Characteristics and examination of 
 deep seated water. 
 
 Chapter VIII. — Artificial Improvement of Natu- 
 ral Water 146-180 
 
 Sedimentation. — Storage. — Aeration. — Filtration. — Principles of 
 sand filtration. — Practical results of artificial filtration. — Sand filtra- 
 tion in the United States. — Advantages of covered filter beds. — Ex- 
 pense. — Other filtering materials. — Household filtration. 
 
 Chapter IX. — Artificial Improvement of Natu- 
 ral Water {continued') 181-193 
 
 Softening of hard water. — Temporary hardness. — Permanent hard- 
 ness. — Chemical processes. — Distillation. 
 
 Chapter X. — Some General Considerations 194-215 
 
 Quantity and waste. — Conduits and distribution pipes. — Service 
 pipes. 
 
 Bibliography 216-220 
 
 Tables for Calculations 221-225 
 
INTRODUCTORY CHAPTER. 
 
 SOLUTION. 
 
 No water which occurs in nature is pure in the strict chem- 
 ical sense of the term, but all natural waters, however free from 
 suspended particles of foreign matter which are visible to the 
 eye, invariably contain in solution more or less of substances 
 which, in their ordinary condition, are solids or gases. It is 
 therefore important, in the beginning, to understand some of 
 the many things which might be said of solution in general. 
 
 V 
 
 Solution of Solid Substances. 
 
 If some pure salt be put into water, after a time the salt dis- 
 appears from sight, and becomes incorporated with the water, so 
 that it is no longer possible to distinguish it by the eye, or to 
 remove it by ever so fine a filter. As far as we can make it out, 
 the change that has taken place is as follows : The ultimate par- 
 ticles of the salt (the molecules of the chemist) are no longer 
 held together in a solid mass by that mutual attraction which we 
 call cohesion, but have become separated from each other and 
 distributed among the particles (molecules) of the water so as to 
 form a homogeneous mixture. As far as we can perceive, it is, 
 indeed, simply a mixture — a mixture of particles of salt with 
 particles of water — we can discover no chemical change, we can 
 trace no chemical action between the dissimilar substances, salt 
 and water. This is an example of what is usually c2iS!ie.di physical 
 solution. The solution differs, of course, essentially from the 
 water. The transparency is not noticeably impaired,* but if 
 
 * This would not be true if, instead of common salt, we had taken a strongly 
 colored substance like the permanganate of potash. In such a case the strong color 
 of the solution would perceptibly diminish its transparency ; otherwise the phenom- 
 ena would be as above. 
 
2 INTRODUCTORY CHAPTER. 
 
 much salt has been used the mobility of the water has been 
 lessened, the boiling point has been raised, the freezing point 
 and the temperature of maximum density have been lowered, 
 the specific gravity, the specific heat and the electrical resistance 
 have also been changed. If, now, the solution be allowed to 
 evaporate at the ordinary temperature, or if the evaporation be 
 hastened by artificially raising the temperature, the water passes 
 off as vapor and the salt rs recovered unchanged. 
 
 If a strip of zinc be immersed in ordinary muriatic (hydro- 
 chloric) acid somewhat diluted, the zinc gradually disappears, 
 but at the same time there is a marked effervescence, due to the 
 escape of hydrogen gas, as well as a considerable increase of 
 temperature. There is, in fact, evidence enough that chemical 
 change is taking place. When the action is over, we have a 
 transparent liquid which is sometimes spoken of as a " solution 
 of zinc," and the phenomenon is spoken of as chemical solution. 
 If, however, we study the action that occurs, we find that it may 
 be regarded as taking place in two steps : in the first place, the 
 zinc acts on the acid to set free hydrogen gas, and to form a new 
 compound, chloride of zinc ; in the second place, the chloride of 
 zinc thus formed dissolves in the liquid present, as did the salt in 
 the previous example. 
 
 Using the term solution in its ordinary sense, to cover all 
 cases of the disappearance of a solid * in a liquid, there are many 
 cases which can be assigned without hesitation to one or the 
 other of the two classes of actions just described, but it is by no 
 means easy in all cases to say whether we are dealing with a 
 simple (physical) solution, or whether chemical action also takes 
 place. Thus, if dry oxide of sodium be put into water, it dis- 
 solves, that is, the solid disappears, but we are quite sure in this 
 case that it is not the oxide of sodium which exists in the solu- 
 tion ; in fact, the oxide of sodium combines chemically with a 
 portion of the water to form hydrate of sodium,f and it is this 
 hydrate, and not the oxide, which actually dissolves in the mass 
 of the water present. Again, if dry carbonate of soda be put 
 into water, there is next to no doubt that it combines with a 
 
 * The solution of gases and liquids will be considered further on. 
 \ The chemical change which takes place is thus symbolized : NaaO + HjO = 
 2NaOH. 
 
SOLUTION OF SOLID SUBSTANCES. 3 
 
 portion of the water, and that the compound which actually 
 enters into solution has the same composition as the crystallized 
 carbonate which contains, in addition to the elements of carbonate 
 of soda, also a quantity of water. With regard to many solids, 
 we are in doubt whether they dissolve simply as such, or first 
 undergo chemical change. In cases like that of common salt, 
 first mentioned, we can trace no such chemical change, but it is 
 by no means certain that no chemical action takes place, and 
 many regard the solution of salt and other similar substances as 
 due to the same cause as that to which chemical action in gen- 
 eral is due, manifested, however, to a slight degree.'" 
 
 The term solution is applied almost exclusively to cases where 
 the dissolving substance, the solvent, is a liquid ; the substance 
 dissolved may be solid, liquid, or gaseous. Every liquid can act 
 as a solvent for certain things, although, if the liquid is of a 
 marked alkaline or acid character, solution is usually accompanied 
 by chemical change. Water is spoken of as the universal sol- 
 vent, and more than any other liquid does it dissolve various 
 substances without evident change. In what follows, we shall 
 use the terms soluble and insoluble to mean soluble and insoluble 
 in water. 
 
 Solids differ from each other very much in the facility with 
 which they may be dissolved in water ; thus, chloride of calcium, 
 if merely left exposed to the air, readily absorbs from the atmos- 
 phere enough water to dissolve it, and is an example of a so- 
 called deliquescent substance ; on the other hand, sulphate of lead 
 requires about 23,000 times its own weight of water (Fresenius), 
 and quartz may be said to be insoluble. At a given temperature 
 a certain quantity of water will always dissolve the same weight 
 of a particular substance, and, as a general rule, the amount 
 which can be dissolved increases with increase of temperature. 
 The effect of increased temperature, however, differs greatly with 
 different substances. Thus chloride of sodium (common salt) is 
 more soluble in hot than in cold water, but only very slightly 
 so ; in the case of chloride of potassium, however, the amount 
 dissolved increases regularly with the temperature, while in the 
 case of nitrate of potassium the increase of solubility for a 
 
 * See. for instance, a paper On Solution and the Chemical Process. Hunt's 
 Chemical and Geological Essays, pp. 44S and foil. 
 
INTRODUCTORY CHAPTER. 
 
 Ftg. 1.— curves of solubility. 
 
 given increase of temperature is greater the higher the tempera- 
 ture : 
 
 DIFF. FOR 
 GRAMS. lo''. 
 
 lOo grams of water at o" C, dissolve of chloride of potassium. 29.23 I 
 
 .70, 
 
 .18 f = 
 
 40 
 
 100 grams of water at o° dissolve of nitrate of potassium . 
 20° 
 40° 
 60° 
 
 34 70 
 40 
 
 45-66 i" 5 
 '3-32 I j8 
 
 31-70 [ 
 63 97 j 
 110.33 \ 
 
 46 
 
 47 
 43 
 48 
 
 38 
 = 7 
 36 
 
SOLUTION OF SOLID SUBSTANCES. 
 
 5 
 
 The diagram on the opposite page shows the relation of solu- 
 bility to temperature in the case of a number of common salts. 
 The horizontal lines indicate the number of parts by weight of 
 the salt which lOO parts by weight of water will dissolve at the 
 indicated temperature. 
 
 There are some remarkable exceptions to the general rule 
 that a larger amount of a substance can be dissolved in a given 
 quantity of hot water than in the same quantity of water at a 
 lower temperature. Sulphate of lime is one of these exceptions. 
 This substance occurs in sea-water and in other natural waters, 
 and, by virtue of the fact that its solubihty decreases with 
 increase of temperature, it gives much trouble in steam-boilers 
 by separating out from the water and forming a coherent scale. 
 The following table shows the solubility of sulphate of lime in 
 sea-water at temperatures above 103° Centigrade. 
 
 TABLE I. — Solubility of Sulphate of Lime in Sea-Water.* 
 
 
 
 
 Sea-water Saturated with 
 
 Sulphate of 
 
 Temperature in Degrees 
 
 
 
 i^IME 
 
 
 Centigrade. 
 
 Fahrenheit. 
 
 Pressure in 
 Atmospheres. 
 
 Marks (at 15** 
 
 C.) onJBeaum^'s 
 
 hydrometer. 
 
 Or has a specific 
 gravity of 
 
 Contains per 
 
 cent of sulphate 
 
 of hme. 
 
 103. 
 
 217.4 
 
 I- 
 
 12°. 5 
 
 i.ogo 
 
 0.500 
 
 103 . 80 
 
 218.8 
 
 I. 
 
 12° 
 
 1.085 
 
 0.477 
 
 105.15 
 
 221.3 
 
 I. 
 
 11° 
 
 1.078 
 
 0.432 
 
 108.60 
 
 227-5 
 
 1.25 
 
 10° 
 
 I 070 
 
 0-395 
 
 HI. 00 
 
 231-8 
 
 1.25 
 
 9° 
 
 1.063 
 
 0.355 
 
 113.20 
 
 235.8 
 
 i.25 
 
 8° 
 
 1.056 
 
 0.310 
 
 115.80 
 
 240.4 
 
 1.50 
 
 7° 
 
 1.048 
 
 0.267 
 
 118.50 
 
 245-3 
 
 1.50 
 
 6° 
 
 1. 041 
 
 0.226 
 
 121.20 * 
 
 250.2 
 
 1.50 
 
 5° 
 
 1.034 
 
 0.183 
 
 124. 
 
 255-2 
 
 2. 
 
 4° 
 
 I 027 
 
 0.140 
 
 127.90 
 
 262.2 
 
 2. 
 
 3° 
 
 1.020 
 
 0.097 
 
 130. 
 
 266.0 
 
 2.50 
 
 2 
 
 1. 013 
 
 0.060 
 
 133-30 
 
 271.9 
 
 2.50 
 
 1° 
 
 1.007 
 
 0.023 
 
 This table shows that when sea-water is boiled under a press- 
 ure of one atmosphere, or at a temperature of 103° C, it will 
 become saturated with sulphate of lime, when, by evaporation, 
 its density has been elevated to 12". 5 Beaum6, and it will then 
 contain 0.5 per cent of the sulphate; at 1.25 atmospheres, or 
 108°. 6 C, the water will be saturated with sulphate when it 
 
 * Couste : Annales des Mines, v (1854), p. 80. The second and fifth columns 
 have been added to Couste"s table. 
 
6 INTRODUCTORY CHAPTER. 
 
 marks iO° B., and will then contain 0.395 per cent of the sul- 
 phate. ' At a pressure of two atmospheres, that is, at a tempera^ 
 ture of about 125" C, sea-water, in its natural state and without 
 having undergone any concentration, is very near the point at 
 which saturation occurs, for the density of sea-water is from 3° 
 to 3^.5 B., and this, according to the table, would correspond to 
 a temperature of about 125" C. Sulphate of lime becomes com- 
 pletely insoluble, either in fresh or sea-water, at temperatures 
 between 140° C. and 150° C. (284°-302° F.). 
 
 Saturated and supersaturated solutions. — At any particular 
 temperature a solvent can take up a definite amount of a sub- 
 stance soluble in it, and no more. The solution is said to be 
 saturated when it contains as much of the dissolved substance as 
 can be taken up at the given temperature. It is, however, often 
 possible to prepare what are called supersaturated solutions, by 
 making a saturated solution at some higher temperature and 
 allowing the liquid to cool. Suppose, for example, that a satu- 
 rated solution of nitrate of potassium was made at 60° using 
 100 c.c. of water ; according to the diagram on page 4, at that 
 temperature 100 grams of water would dissolve 1 10-33 grams of 
 the salt. If, now, the solution were cooled to 20°, we should 
 expect that there would then remain dissolved 3 1 . 70 grams, and 
 that 78.63 grams would separate in the solid condition during 
 the cooling; in this particular case, the expectation would be 
 approximately realized ; but, with many salts, if a saturated solu- 
 tion be cooled quietly without agitation, it happens that, when 
 cold, the liquid still retains more of the salt than it could take up 
 at the lower temperature. In fact, Storer says : * "It is often 
 exceedingly difBcult thus to obtain normally saturated solutions, 
 even of our most common and easily crystallized salts, within the 
 limits of time which can be conveniently allotted to a single 
 experiment." 
 
 In the case of some particular salts the phenomenon of super- 
 saturation is very marked. Thus, at a temperature of 33° C, 
 water will dissolve about half its own weight of Glauber's salt 
 (hydrated sulphate of sodium), and if the solution be protected 
 from dust and allowed to cool quietly to the ordinary tempera- 
 ture, the whole of the compound remains in solu'iion. If, now, 
 
 * Dictionary of Solubilities, Preface. 
 
SOLUTION OF SOLID SUBSTANCES. 7 
 
 a small crystal of the same kind as the original salt be dropped 
 into the solution, about three-fifths of the salt separates out in 
 the crystalline form, and the whole mass becomes nearly solid. 
 
 Condition of mixed solutions. — Natural waters usually contain 
 a variety of foreign substances, and it is worth while, at this 
 point, to inquire into the condition of things which exist in a 
 dilute solution containing at the same time several different salts. 
 We will first take a simple case where chemical analysis shows 
 the existence of, say a chloride and a sulphate, and of potassium 
 and sodium. The question may be asked, whether the solution 
 contains chloride of sodium and sulphate of potassium, or sul- 
 phate of sodium and chloride of potassium, or whether all four 
 of these compounds exist at the same time. Now, while the 
 latter is believed to be the true state of the case, chemical 
 analysis is powerless to answer the question, and two solutions 
 undistinguishable from each other could be prepared by taking 
 each pair of salts indicated above in the right proportion.* If a 
 number of soluble salts are mixed together in solution, the matter 
 becomes more complicated, but in stating the results of analysis, 
 certain conventional forms of statement are adopted. . Thus we 
 might mix together in dilute solution and in the right proportion, 
 sulphate of sodium and chloride of calcium, but if the solution 
 were analyzed, it would be reported as containing sulphate of 
 calcium and chloride of sodium, for the reason that the sulphate 
 of calcium is much less readily soluble in water than the other 
 compounds, and if the solution in question were gradually con- 
 centrated by evaporation, the sulphate of calcium would separate 
 in the solid form ; if the evaporation were carried to dryness, the 
 sulphate of calcium and chloride of sodium would remain, each 
 crystallized by itself. For this reason chloride of calcium would 
 never be represented as existing in the presence of sulphate of 
 sodium, although, in a dilute solution, it is probable that a por- 
 tion of the calcium would actually exist in the form of chloride. 
 It should be said that much difference of opinion and practice 
 exists with reference to the best way to represent the constit- 
 uents of a mixed solution, and two chemists will often report 
 
 * We are not in absolute ignorance as to the laws which govern the partition of a 
 base among different acids, or of an acid among different bases, or, indeed, of the acid 
 and basic radicals among diflferent salts. Our knowledge, however, is too limited to be 
 applied practically to any considerable extent in such a case as that of a natural water. 
 
INTRODUCTORY CHAPTER. 
 
 very different statements of the analysis of the same solution, 
 while the numerical results actually obtained are the same. 
 
 The following statement of the analysis of Croton water, by 
 Professor Chandler, is an example of the form in which such 
 reports are usually made : 
 
 Solids Contained in one Gallon of Croton Water. 
 
 Chloride of Sodium 
 
 Sulphate of Potash 
 
 Sulphate of Soda . . . . T. . . . 
 
 Sulphate of Lime 
 
 Bicarbonate of Lime 
 
 Bicarbonate of Magnesia. . 
 
 Silica 
 
 Alumina and Oxide of Iron 
 Organic Matter 
 
 Total 
 
 6.873 
 
 Summer, 1869. 
 
 May II, 1872. 
 
 Grains. 
 
 Grains. 
 
 0.402 
 
 0.284 
 
 0.179 
 
 0.205 
 
 0.260 
 
 0.024 
 
 0.158 
 
 0.024 
 
 2.670 
 
 2.331 
 
 1.913 
 
 1.338 
 
 0.621 
 
 0.222 
 
 a trace 
 
 058 
 
 0.670 
 
 0.874 
 
 5.360 
 
 The following statement represents the actual results of an- 
 alysis, from which the previous statement was made up accord- 
 ing to the conventional plan usually adopted : 
 
 Solids Contained in one Gallon of Croton Water. 
 
 Summer, l86g^ 
 Grains. 
 
 May II, 1872. 
 Grains. 
 
 Soda 
 
 Potash 
 
 Lime .«...." 
 
 Magnesia 
 
 Chlorine 
 
 Sulphuric Acid 
 
 Silica 
 
 Alumina and Oxide of Iron 
 
 Carbonic Acid (calculated) 
 
 Water in Bicarbonates (calculated) 
 
 Organic and Volatile Matter 
 
 Less Oxygen equivalent to the Chlorine 
 Total 
 
 0.326 
 0.097 
 0.988 
 0.524 
 0.243 
 0.322 
 0.621 
 a trace 
 2.604 
 0.532 
 0.670 
 
 6.927 
 0.054 
 
 6.873 
 
 0-157 
 o.iog 
 0.819 
 0.369 
 0.172 
 0.124 
 0.222 
 0.058 
 2.074 
 0.421 
 U.874 
 
 5-399 
 0.039 
 
 5.360 
 
 Effect of the presence of one substance on the solubility of another 
 substance. — When water has dissolved as much of a given sub- 
 stance as it can under the existing conditions, it is still able to 
 take up more or less of a second substance. In general, the 
 
SOLUTION OF SOLID SUBSTANCES. 
 
 9 
 
 presence of one substance in a solution will modify the action of 
 the liquid on another substance presented to it ; thus, while 
 cyanide of silver is insoluble in water, it dissolves readily in an 
 aqueous solution of cyanide of potassium. Again, although car- 
 bonate of lime dissolves only to a very slight extent in pure 
 water, it dissolves to a considerable extent in water containing 
 carbonic acid, and is a frequent constituent of natural waters. 
 When such waters are boiled, the carbonic acid escapes, and as 
 the water is no longer able to keep the carbonate of lime in solu- 
 tion, this compound separates out and is one of the substances 
 which cause trouble in steam boilers by forming an incrustation 
 or scale. Undoubtedly, in such cases as those mentioned, chem- 
 ical action takes place to a greater or less extent between the 
 substances dissolved. 
 
 Various ineans of hastening solution. — The solution of a solid 
 substance is facilitated by any means which tends to bring con- 
 tinually fresh portions of the solvent into intimate contact with 
 the substance to be dissolved. It is consequently of advantage 
 to. reduce the solid to a fine powder, and to agitate the mixture. 
 Where there is no objection to heating the liquid, the solution is 
 facilitated by so doing. Advantage may be taken of the fact 
 that the solution has a greater specific gravity than the solvent, 
 by suspending the solid near the sur- 
 face of the liquid. This may be 
 easily shown by putting crystals of 
 some colored salt, as bichromate of 
 potash or sulphate of copper, into a 
 cone of wire gauze, or into a funnel, 
 or into a tube over one end of which 
 a piece of cloth is tied, and suspend- 
 ing the apparatus in a vessel of water. 
 The solution, as it is formed, falls 
 toward the bottom of the vessel, 
 marking its path through the water 
 by colored threads, and this action continues until the solution is 
 saturated. 
 
 In preparing brine or a solution of copperas — as a disinfectant, 
 for instance — much time may be saved by suspending the salt or 
 the copperas in a basket or coarse bag near the top of the tub or 
 barrel in which the solution is made, instead of putting the solid 
 
 Fig. 2. 
 
lO INTRODUCTORY CHAPTER. 
 
 at the bottom of the water and endeavoring to hasten solution 
 by stirring. 
 
 Effect of dissolved solids on various physical phenomena. — The 
 effect of dissolved sohd substances is to increase the specific 
 gravity, and in the case of many substances tables have been 
 prepared from which the percentage of the substance dissolved 
 can be learned by observing the specific gravity of the solution. 
 Similar tables have also been prepared for solutions of gases and 
 liquids. 
 
 The presence of dissolved substances raises the boiling point 
 of the solution, and the stronger the solution, the higher the 
 boiling point. Thus, Ordinary sea-water boils at about ioi° C. 
 (2 1 3". 8 F.), while a saturated solution of acetate of potash does 
 not boil until the temperature reaches 169° C. (304". 2 F.). 
 
 The effect of dissolved solids is to lower the freezing point 
 and the temperature of maximum density. Thus, while pure 
 water is at its maximum density at 4° C. (39° . 2 F.), and freezes 
 at 0° C. (32° F.), sea-water, according to Despretz, acquires its 
 maximum density at 3". 67 C. (38°. 6 F.), and freezes at — i^-BS C. 
 (28°. 6 F.). 
 
 Solution of Gases. 
 
 It is true of gases as of solids that they vary very much in 
 their deportment toward water, some being absorbed by it with 
 very great readiness, while others may be kept in contact with it 
 for a long time without suffering any considerable decrease of 
 bulk. It is generally true, however, of gases, that they dissolve 
 to a certain, even if to a slight, extent, and the amount is fixed 
 and definite under the same conditions of temperature and press- 
 ure. As a rule, increase of temperature decreases the solubility, 
 and by prolonged boiling, water may be freed from the gases 
 which it holds in solution.* The specific gravity of solutions of 
 gases is sometimes lower and sometimes higher than that of water. 
 
 The volume of a gas (expressed in cubic centimeters and 
 measured at 0° C, and under a pressure of 760 m.m. of mercury) 
 which one cubic centimeter of water will dissolve at a certain 
 
 * This is not invariably true. Thus, a strong solution of chlorhydric acid gas, 
 when heated, gives oiT the gas until the amount of acid in the remaining solution has 
 been reduced to 20.24 per cent by weight; the solution of this strength distills 
 unchanged at 110° C. 
 
SOLUTION OF GASES. 
 
 II 
 
 temperature, is called the coefficient of absorption at that tem- 
 perature. The accompanying table gives the coefficient of absorp- 
 tion at various temperatures for oxygen, nitrogen, carbonic acid, 
 and ammonia.* 
 
 TABLE II. — Coefficient of Absorption of various Gases. 
 
 Temperature. 
 
 Oxygen. 
 
 Nitrogen. 
 
 Carbonic Acid. 
 
 Ammonia. 
 
 o° C. 
 
 0.041 14 
 
 0.02035 
 
 1.7967 
 
 1049 . 6 
 
 I 
 
 0.04007 
 
 O.OI981 
 
 1 . 7207 
 
 1020.8 
 
 2 
 
 . 03907 
 
 0.01932 
 
 I. 6481 
 
 993-3 
 
 3 
 
 0.03810 
 
 0.01884 
 
 1-5787 
 
 967.0 
 
 4 
 
 0.03717 
 
 0.01838 
 
 I. 5126 
 
 941.9 
 
 5 
 
 0,03628 
 
 0.01794 
 
 1-4497 
 
 917.9 
 
 6 
 
 0.03544 
 
 0.01752 
 
 I • 3901 
 
 895.0 
 
 7 
 
 0.03465 
 
 0.01713 
 
 1-3339 
 
 873 -.1 
 
 8 
 
 0.03389 
 
 0.01675 
 
 I . 2809 
 
 852.1 
 
 9 
 
 0.03317 
 
 u. 01640 
 
 1-2311 
 
 832.0 
 
 lO 
 
 0.03250 
 
 0.01607 
 
 1.1847 
 
 812.8 
 
 II 
 
 0.03189 
 
 0.01577 
 
 I. 1416 
 
 794-3 
 
 12 
 
 0.03133 
 
 0.01549 
 
 I. 1018 
 
 776-3 
 
 13 
 
 0.03082 
 
 0.01523 
 
 1-0653 
 
 759-6 
 
 14 
 
 . 03034 
 
 0.01500 
 
 I. 0321 
 
 743-1 
 
 15 
 
 0.02989 
 
 OI47S 
 
 1 . 0020 
 
 727.2 
 
 i6 
 
 . 02949 
 
 0.0145S 
 
 0-9753 
 
 711. 8 
 
 17 
 
 0.02914 
 
 0.01441 
 
 0.9519 
 
 696.9 
 
 i8 
 
 0.02884 
 
 0.01426 
 
 0.9318 
 
 632.3 
 
 19 
 
 0.02858 
 
 O.OI4I3 
 
 0.9150 
 
 668.0 
 
 20 
 
 0.02838 
 
 0.01403 
 
 0.90J4 
 
 654.0 
 
 From the table it is seen that the coefficient of carbonic acid 
 is considerably, and that of ammonia enormously, greater than 
 that of oxygen and nitrogen. In the case of gases like ammonia, 
 hydrochloric acid, etc., we have no difficulty in believing that the 
 great absorption is due to the fact that chemical combination 
 takes place, and this is probably true also of carbonic acid, sul- 
 phuretted hydrogen, etc. It is a curious fact that, with hydrogen, 
 the coefficient of absorption in water is the same at- all tempera- 
 tures from 0° to 20° C. 
 
 If the pressure be increased, a greater weight of the gas is 
 dissolved at a given temperature, and the increase is, in fact, 
 proportional to the increase of pressure, but since the increased 
 pressure diminishes the volume of the gas in the same propor- 
 tion, the same actual volume of gas is dissolved whatever the 
 pressure. In the case of gases which are easily liquefied by press- 
 ure, or of those which are very soluble in water, this law does 
 not hold good under all circumstances. Thus carbonic acid fol- 
 
 * Bunsen : Gasometrische Methoden, Braunschweig, 1877. 
 
12 INTRODUCTORY CHAPTER. 
 
 lows the law only when the pressure is small, and ammonia gas 
 only at high temperature. The law is, in general, true also when 
 water is exposed to an atmosphere of mixed gases ; that is, each 
 of the gases in the mixture is dissolved according to its own 
 coefificient of solubility ; but in determining the actual quantity 
 of any one gas dissolved {i. e., the volume at o° and 760 m.m.), it 
 must be remembered that the pressure exerted by the mixed 
 gases is made up of the partial pressures of the several gases in 
 the mixture. Thus, whether water at 0° C. be exposed to pure 
 oxygen with a pressure of 760 m.m., or to air at the pressure of 
 760 m.m., the same volume of oxygen will be dissolved in either 
 case, namely, as we see from the table, 0.041 14 c.c. of oxygen in 
 each cubic centimeter of water ; but in the one case we have 
 0.041 14 c.c. of oxygen with a pressure of 760 m.m., whereas in 
 the other case we have 0.041 14 c.c. of oxygen with a pressure of 
 one-fifth of 760 m.m., for in the air the oxygen forms only about 
 one-fifth part by volume. It follows that the actual quantity of 
 gas {i. e., the volume which it would have at 0° and 760 m.m.) in 
 the first case is about five times as great as in the second. It is 
 possible; knowing the composition of the mixture of dissolved 
 gases in any case, to calculate the composition of the atmosphere 
 to which the water was exposed, but this does not hold in the- 
 case of a gas forming only a very small part of a mixture. 
 Owing to the fact that the coefificient of absorption for oxygen is 
 greater than that of nitrogen, water which has been exposed to 
 ordinary air contains these two gases in quite a different propor- 
 tion from that in which they exist in the air, and this is one of 
 the proofs adduced to show that in the air the gases were simply 
 mixed together, and not chemically combined. The relation is 
 as follows : 
 
 Composition of Air Composition of " Dissolved 
 by Volume. Air" by Volume.* 
 
 Oxygen -. 20.96 34-91 
 
 Nitrogen 79-04 65.09 
 
 Supersaturated solutions of gases. — As in the case of solids, it 
 is possible to prepare temporarily supersaturated solutions of 
 gases. Thus, if ordinary " soda-water " be drawn from a siphon 
 or from a fountain, a quantity of gas escapes as soon as the 
 
 * Bunsen : Gasometrische Methoden, p. 224. 
 
SOLUTION or GASES. 1 3 
 
 excess of pressure is removed. Afterward the gas escapes more 
 slowly, until eventually the water contains no more carbonic 
 acid gas than it would take up under the conditions to which it 
 is now exposed. That we really are dealing with a supersaturated 
 solution after the first rapid escape of gas, may be shown by 
 dropping into the soda-water a teaspoonful of sugar, or some 
 sand, or almost any powder ; this causes immediate effervescence 
 and escape of gas by furnishing free surfaces and sharp angles for 
 the collection and liberation of the gas within the liquid. 
 
 Facilitating gaseous solution. — In preparing a saturated solu- 
 tion of a gas, the latter is generally allowed to simply bubble 
 through the liquid to be saturated until the end is attained, the 
 liquid being agitated or not, according to convenience. In the 
 case of most gases, solution is facilitated by lowering the tem- 
 perature, as, for instance, by surrounding the vessel in which the 
 absorption takes place with cold water or ice. This is of especial 
 use when the gas enters into chemical combination with the 
 liquid, as is true of ammonia, hydrochloric acid, etc., when dis- 
 solved in water, because the heat of chemical action raises the 
 temperature of the solution. 
 
 It is sometimes desirable to charge water with gas under 
 increased pressure ; thus water is charged under pressure with 
 carbonic acid gas, to bear thenceforth the name of " soda-water ; " 
 sulphuretted hydrogen water, for use in chemical laboratories, is 
 also prepared similarly. 
 
 Solubility of Liquids in Water. 
 
 Like solid substances, liquids differ among themselves as to 
 their solubility in water, some liquids, as glycerine and common 
 alcohol, mix with water in any and all proportions ; others, like 
 ether and amylic alcohol, dissolve to a limited and definite extent ; 
 others, like some oils and mercury, seem to be altogether in- 
 soluble, although by long contact with the water such insoluble 
 liquids may undergo chemical change, and give up something to 
 the water. It is more difficult than with solids to determine 
 whether and to what extent an insoluble or difficultly soluble 
 liquid dissolves in water, and to distinguish between solution and 
 suspension, especially if the liquid is colorless and has nearly the 
 same refractive index as water. In this case it is impossible to 
 
14 
 
 INTRODUCTORY CHAPTER. 
 
 obtain satisfactory indications with the eye, and complete separa- 
 tion of suspended particles is often impossible. The solution of 
 a difificultly soluble colorless liquid may sometimes be watched 
 by coloring the liquid with some substance which does not inter- 
 fere with the action. Thus, if amylic alcohol be colored with a 
 little iodine, the process of solution may be followed more readily 
 than would otherwise be possible. 
 
 As in the case of solids and gases, we have the same grada- 
 tion from cases of marked chemical action, accompanied by dis- 
 play of heat and decrease of bulk, to cases where we seem to be 
 dealing with a simple mechanical mixture. 
 
 Distinction between Solution and Suspension. 
 
 There is a great deal of confusion, even among well-educated 
 people, as to the proper use of the terms in solution and in suspen- 
 sion, and in the same connection it may be said that a great deal 
 of confusion exists with reference to the distinction between 
 clear and colorless, terms which are by no means synonymous. 
 The accurate use of the terms can probably best be made plain 
 by illustraitions. If, to take an example already made use of, we 
 put some common salt into a quantity of water, after a time the 
 salt disappears, the ultimate particles being distributed through 
 the water so that they are no longer distinguishable by the eye, 
 even aided by the most powerful microscope ; the salt cannot be 
 removed by simple filtration ; and, although the solution is some- 
 what less mobile than water, it is still transparent. This, as we 
 have seen, is a case of solution. Suppose that, instead of the 
 salt, we take a quantity of sulphate of copper (blue vitriol). The 
 phenomena will be similar, but the blue color of the compound 
 shows itself in the solution. The more concentrated the solu- 
 tion, the more will its transparency be diminished on account 
 of the depth of color ; it is easy, however, by taking a thin layer 
 of the solution to satisfy one's self of the transparency of the 
 liquid and of the absence of suspended particles. Such a liquid, 
 although colored, is clear. 
 
 Suppose, now, we take some clay, shake it with water, and 
 then allow it to settle. The grosser particles will subside to the 
 bottom of the vessel, but the finer particles will remain in sus- 
 pension. Very finely divided clay will refuse to settle for weeks, 
 
DISTINCTION BETWEEN SOLUTION AND SUSPENSION. 1 5 
 
 and sometimes even for months. In such cases the liquid appears 
 somewhat turbid and opaque ; and, although the individual par- 
 ticles are too fine to be readily removed by ordinary filters and 
 too small to be distinguished as particles by the eye, still the 
 clay has not dissolved, and the very turbidity or opacity of the 
 liquid shows the presence of solid particles, although they are 
 extremely minute. Such an appearance is not to be described as 
 "being colored," although finely divided clay and other material 
 may be suspended in a liquid which does of itself possess a dis- 
 tinct color. One often meets with the expression, and that, too, 
 in standard works, " the water is discolored by clay," when really 
 it is a question of a colorless water carrying particles of clay in 
 suspension. As we shall see further on, surface waters are often 
 highly colored by vegetable extractive matter in solution, but 
 the water may at the same time be perfectly clear and transpa- 
 rent. On the other hand, pond waters often appear decidedly 
 green; but simple filtration gives a colorless water, and shows the 
 green color to have been due to particles of green (vegetable) 
 matter which were suspended in the liquid. 
 
 While for practical purposes there is no difficulty in distin- 
 guishing that which is really dissolved from that which is merely 
 held suspended, and in applying the terms as already indicated, 
 it is true that there are substances, generally considered insoluble, 
 which admit of such minute subdivision that the finer particles 
 will remain suspended in water for months, giving in some cases 
 a faint opalescence to the liquid, but in other cases apparently 
 only a color. Thus, by trituration with milk-sugar, metallic gold 
 may be reduced to so fine a condition that it will diffuse through 
 water, giving it a purple color, and it is hard to say that the gold 
 does not exist in solution; after long standing, however, the 
 metal separates out and settles to the bottom of the liquid, and 
 this separation may be hastened by the addition of certain saline 
 solutions.* The action of the saline solution is not fully under- 
 stood, but it was noticed long ago that these minute particles 
 showed under the microscope the so-called Brownian motion, and 
 that the addition of small quantities of alum, glue, lime, car- 
 bonate of ammonia or other salts, caused this molecular motion 
 
 * Buchmann : Beobachtungen und Untersuchungen zum Nachweis der Loslich- 
 keit von Metallen, etc. Leipzig, i8Si, p. 54. 
 
1 6 INTRODUCTORY CHAPTER. 
 
 to cease, and the particles to flock together and settle out as 
 minute, amorphous, curdy masses.* The same thing is illus- 
 trated on the large scale by the phenomena which take place 
 when a river, like the Mississippi, loaded with silt, meets the 
 waters of the ocean. Dr. Hunt f found that water taken near 
 the mouth of the Mississippi contained about -j-oVir of suspended 
 matter, mainly clay, which required from ten to fourteen days to 
 subside. He, however, observed that the addition of sea-water 
 or of salt, sulphate of magnesia, alum, or sulphuric acid, rendered 
 the water clear in from twelve to eighteen hours, owing to this 
 same flocculation of the suspended matter. 
 
 * Naumann : Thermochemie, p. 33. 
 
 t Proc. Boston Soc. Nat. Hist., xvi, p. 301. 
 
WATER SUPPLY. 
 
 CHAPTER I. 
 
 DRINKING WATER AND DISEASE. 
 
 With reference to their use for town and household supply, 
 we shall roughly divide all natural waters into four classes, as 
 follows : 
 
 1. Rain water ; 
 
 2. Surface water, including streams and lakes ; 
 
 3. Ground water, including shallow wells ; 
 
 4. Deep-seated water, including deep wells, artesian wells, and 
 springs. 
 
 Under each of these heads we shall study the advantages and 
 disadvantages of the particular class of water, the hability of pol- 
 lution, etc. ; but first we will consider, in a general way, the con- 
 nection which exists, or is supposed by some to exist, between 
 drinking water and disease. 
 
 A water containing a considerable amount of dissolved sub- 
 stances^^— one which could properly be denominated a mineral 
 water — would not be thought of for a public water supply, and 
 would seldom be used as a regular beverage except for the sake 
 of real or fancied medicinal effect ; a small amount, however, of 
 mineral matter is generally considered an advantage. The pres- 
 ence of the substances which ordinarily exist in solution in natu- 
 ral waters must not be regarded as necessary, because experience 
 on ship-board has shown that distilled water, properly aerated, is 
 perfectly wholesome. It appears that distilled water, soft sur- 
 face water, and moderately hard* spring or well water are all 
 wholesome, and may be drunk without inconvenience by persons 
 accustomed to their use. It is, however, true that a person who 
 
 * A /lartf water is, generally speaking, one which contains compounds of lime or 
 magnesia in solution. See pages 33 and 181. 
 
1 8 WATER SUPPLY. 
 
 is in the habit of drinking a soft water generally experiences 
 some derangement of the digestive organs on beginning to use 
 hard water, and vice versa. It is contended by some that the 
 human system needs salts of lime, etc., that these compounds are 
 furnished in an assimilable form in water, and that, consequently, 
 a somewhat hard water is more advantageous for town supply ; 
 statistics have been brought together to support this view by 
 comparing the death rate of various towns with the hardness of 
 the water supply, but the death rate depends upon too m_any 
 factors to be used as the chief argument in this connection. It 
 is, however, the result of general observation that a hard water 
 pf which the hardness is due to salts of magnesia or to sulphate 
 of lime is not well suited for drinking, and is injurious to most 
 persons. 
 
 Waters, especially surface waters, containing much vegetable 
 matter are also, in some cases, unwholesome. The water of 
 marshes is sometimes the cause of diarrhoea and other diseases 
 of this character, and is supposed by some to cause malarial and 
 other fevers (see also page loo). The mere presence of vegetable 
 organic matter, however, is not sufficient to produce these effects, 
 because many waters which are quite deeply colored by vegetable 
 matter are proved by experience to be perfectly wholesome.* 
 
 While some waters are thus, in their natural condition, 
 unwholesome and may be the cause of sickness, the attention of 
 sanitarians and water experts is directed nowadays principally 
 to the effect of water which is polluted by the waste materials 
 from manufactories and dwellings, or by the sewage of towns 
 and cities ; and it is generally held, especially in England and the 
 United States, that water thus polluted may be, and frequently 
 is, the cause of certain specific diseases. Before discussing this 
 question directly, it is important to have a general idea of the 
 present prevailing view with reference to the so-called zymotic 
 diseases, and to understand what is meant by the " germ theory." 
 
 Many clear liquids containing organic matter of animal or 
 vegetable origin — such, for instance, as infusions of hay, infusion 
 of turnip, urine, etc., etc.,- — if exposed to the air gradually be- 
 come turbid or cloudy, or, perhaps, a film forms on the surface 
 of the liquid, or a deposit upon the walls of the vessel which con- 
 
 * See, however, a remark by Professor Mallet, page 26. 
 
DRINKING WATER AND DISEASE. 19 
 
 tains it. The cause of the turbidity is shown by the microscope 
 to be the presence of countless minute organized bodies — some 
 rod-like, others globular — which prove to be capable of self-prop- 
 agation, and which are endowed with motion, at least under cer- 
 tain conditions. 'Similar organisms are found in the "dust" 
 which floats in the air, and which may be collected by causing a 
 current of air to impinge upon a surface moistened with glycer- 
 ine ; they occur in rain water, particularly in that which falls in 
 the beginning of a shower, in surface waters and elsewhere. They 
 are found especially where there is decomposing organic matter, 
 and perform an active part in promoting or producing the chem- 
 ical changes which take place. In certain diseases of men and of 
 the lower animals, organisms which, in their general character, 
 are similar to those thus described have been found in the blood 
 or in the substance of various organs, and their connection with 
 the disease seems to be something more than a coincidence ; 
 there seems, indeed, to be a causal connection. 
 
 The micro-organisms with which we are now concerned are 
 referred to the vegetable kingdom ; they are regarded as related 
 both to the fiingi and to the algce, and are designated scientifi- 
 cally as schizomycetes (Spaltpilze, Nageli) ;* their study requires 
 the highest powers of the microscope and the greatest skill in 
 observation. The development of certain forms has been care- 
 fully studied, and it is known that they multiply not only by 
 fission — as Nageli's classification implies — but also by the forma- 
 tion of spores or permanent " germs." The " germ theory " of 
 •disease is that many diseases are due to the presence and propa- 
 gation in the system of these minute organisms, which are popu- 
 larly spoken of under the general name bacteria, under which 
 term are included also organisms which, as far as known, are 
 harmless. Some of the diseases which have, with more or less 
 show of reason, been supposed to have their cause in such organ- 
 isms are malarial (intermittent) fever, relapsing fever, typhus and 
 typhoid, cholera, yellow fever, diphtheria, and tuberculosis. 
 
 * Nageli : Die niederen Pilze in ihren Beziehungen zu den Infectionskrankheiten, 
 Munich, 1877. Nageli distinguishes three natural groups among the lower fungi : 
 (i) the mucorini, Schimmelpilze, mould fungi ; (2) the saccharomycetes, Sprosspilze, 
 budding fungi ; (3) the schizomycetes, Spaltpilze, fission fungi. The second class 
 includes the organisms which produce the fermentation of wine and beer ; the third 
 class includes the fungi of putrefaction, the so-called bacteria. The terms microbes, 
 microzymes, as well as several others, are also used to include all these micro-organisms. 
 
20 
 
 WATER SUPPLY. 
 
 With reference to specific distinctions among the organisms 
 themselves observers are not agreed. Some would very much 
 restrict the number of true species, and refer the differences in 
 appearance and action to differences in the attendant conditions ; 
 others believe that there are many species, as distinct as those 
 observed in higher organisms, and that each disease has its own 
 
 bacterium ; they believe 
 that the observed differ- 
 ences are essential, and the 
 inability to recognize, in 
 all cases, satisfactory spe- 
 cific characters is due to the 
 Fig. 3. a. Micrococcus prodigiosus ; b. the same imperfection of the means 
 in the zooglcea stage ; .. M. Ure^. 650 : i. ^j . observation. For pur- 
 
 poses of study, at any rate, the various observed forms may be 
 classified in genera and species. Referring to a few terms of 
 
 .'^% 
 
 * : 
 
 * ? 
 
 C 
 
 4%°o •' 
 
 i^.*' ■ 
 
 Fig 4. li. Bacterium termo ; h. Same in the zoogloea stage ; c. Bacterium lineola ; 
 d. Vibrio rugula ; c, V. serpens j f. Spirillum volutans ; g. Sp. tenue. 650 : 1. 
 
 somewhat common occurrence, the micrococci are globular or 
 oval, the bacilli are rod-like, the vibriones (vibrios) are sinuous, 
 and the spirilla are spiral. No attempt to represent these mi- 
 nute organisms by cuts can be very successful. Figures 3 and 4 
 may serve to give a rough idea of the general appearance of 
 some of them. 
 
 The connection of the bacteria with disease has been most 
 satisfactorily made out in the case of splenic fever (anthrax, 
 charbon (Fr.), Milzbrand (Germ.), malignant pustule). In this 
 disease the blood and various organs, especially the spleen, con- 
 tain an organism known as Bacillus anthracis. Koch* cultivated 
 * Cohn's Beitrage zur Biologie der Pflanzen, ii. p. 277. 
 
DRINKING WATER AND DISEASE. 21 
 
 the organisms in appropriate fluids outside of the animal body, and 
 observed the development and the formation of spores, and from 
 these spores he reproduced the specific disease in living animals. 
 He found that the organisms themselves, as observed in the 
 blood, usually died in a few days, but that the spores retained 
 their vitality for at least four years. The Bacillus anthracis in 
 appearance is scarcely to be distinguished from the Bacillus sub- 
 tilis, a harmless form which occurs in infusions of hay, and Dr. 
 Buchner* has been able, as he claims, by a series of cultivations 
 to transform one form into the other, and from the harmless 
 Bacillus subtilis to obtain the Bacillus anthracis, and to prove its 
 identity by producing the well-marked disease in animals. Nat- 
 urally enough Buchner's tonclusions are disputed, and, until his 
 results are generally accepted by competent specialists, they 
 cannot be looked upon by the world in general as showing more 
 than a possibility. 
 
 Admitting the necessary presence of these minute organisms in 
 the bodies of persons sick with certain diseases, organisms which, 
 at least in certain stages of their development, can exist outside 
 the human body and retain their vitahty for a long time, the 
 question arises how they can find their way into the systems of 
 healthy persons to produce disease. The two most obvious of 
 the possible carriers of disease are the air we breathe and the 
 water we drink. We have no difficulty in supposing that ema- 
 nations from sick persons, particulate or otherwise, may find 
 their way into the air ; moreover, the dejections of the sick and 
 the water in which their clothes or their persons have been 
 washed may often reach wells or other sources of drinking water. 
 Of these two media the former, i. e. the air, is a priori the most 
 probable, partly because we take very much more air into our 
 lungs than we take water into our stomachs, and also because 
 the lungs afford a better chance for the organisms to enter the 
 blood; indeed, some maintain that any organisms entering the 
 stomach are rendered harmless by the fluids therein, and that 
 the drinking water is not to be considered at all as a means of 
 conveying the germs of disease. 
 
 Of the diseases which are supposed to be caused by these 
 micro-organisms — to be propagated by germs — those which have 
 
 * Sitzb. d. math.-phys. Classe d. kanigl. bayer. Akad., 1880, p. 368. 
 
22 WATER SUPPLY. 
 
 been, with the greatest unanimity, ascribed to the use of impure 
 drinking water are typhoid fever and cholera. With reference 
 even to these diseases, however, there has been much discussion 
 and controversy between the adherents and the opponents of the 
 "drinking-water theory" since 1848, when Snow, Budd, and 
 others in England ascribed the spread of the cholera then prev- 
 alent to the drinking of water fouled by the dejections of cholera 
 patients. It would be unprofitable, in brief space, to attempt to 
 review the numerous cases on record where the coincidences 
 between impure water and cholera (and other diseases) have 
 been so marked as to lead able and careful investigators to believe 
 in the existence of cause and effect. The most able opponent 
 of the theory is Professor Pettenkofer, of Munich, who holds 
 that in these cases there is coincidence only, and that other cir- 
 cumstances, and notably the character of the ground and the 
 condition of ground water, have been overlooked in the investi- 
 gations. He and his sympathizers also bring forward many 
 instances where the connection between a particular outbreak of 
 a speciiic disease and the drinking water previously used by those 
 attacked is not only not obvious, but absolutely out of the 
 question. 
 
 Even the most earnest advocates of the drinking-water theory 
 must admit that the theory is by ho meang proved, in the sense 
 in which a mathematical proposition may be proved, and it cer- 
 tainly cannot be asserted that the drinking water is the only 
 means by which the zymotic diseases may be propagated ; the 
 coincidences, however, if coincidences they be, are most remark- 
 able, and every year adds to their number. To choose a single 
 example from the multitude which are accessible, we may take 
 the case of an outbreak of typhoid fever in North Boston, Erie 
 Co., N. Y., which occurred in the year 1843, consequently before 
 the connection between typhoid and drinking water had become 
 a theory. The community consisted of nine families (forty-three 
 persons), and typhoid fever had never been known in the vicinity 
 until in the year named a sick traveler took lodging at the tavern, 
 and twenty-eight days thereafter died of what was pronounced 
 by the physicians to be typhoid fever. The disease spread, twen- ■ 
 ty-eight persons in all were attacked, and only three families es- 
 caped. These three families were the only ones which did not use 
 the well of the tavern, two of them on account of distance, and one 
 
DRINKING WATER AND DISEASE. 23 
 
 on account of a feud between them and the inn-keeper. The 
 latter family lived within four rods of the tavern. The physi- 
 cians, at that time, ascribed the communication of the disease to 
 a " contagium contained in the emanations from the body," but 
 the people charged the family referred to with poisoning the well, 
 so marked was the coincidence. Subsequently, Dr. Flint, in 
 reviewing the case, regards it as " vastly probable, if not certain," 
 that the disease was communicated by the drinking water, the 
 source of which was so situated that it must, in all probability, 
 have been contaminated by the dejections of the first patient.* 
 
 Many other instances might be cited where a number of fam- 
 ilies or persons using the same well or other source of water 
 supply have become victims to the disease, which does not, at 
 the time, appear elsewhere ; there are many instances where the. 
 closing of the suspected source of supply has at once put a stop 
 to the further spread of the disease; there are also instances 
 where people have assembled in numbers on account of some 
 celebration, and sickness has followed in the case of a large pro- 
 portion of those who have used a certain water, while the others 
 have not been affected; latterly there have been cases where 
 sickness has broken out among families obtaining their milk from 
 the same source, and investigation has shown that impure water 
 was used in the dairy. The most weighty criticism urged against 
 the acceptance of the theory which, in such cases as those men- 
 tioned, regards the water as the cause of disease, is that, in the 
 investigations made, sufficient attention has not been paid to 
 other, features of the circumstances and surroundings which 
 might, with equal likelihood, be factors in the production of dis- 
 ease. No doubt, in some instances this criticism is just ; but if 
 we rule out all cases where the observations are manifestly im- 
 perfect there still remain instances enough to make the connec- 
 tion between the water drank and the disease contracted ex- 
 tremely probable. As this is a matter which, in the present state 
 of science, cannot be absolutely proved or disproved, the duty 
 of those who have to advise or to decide in matters relating to 
 water supply is perfectly clear ; it is to err on the side of safety, 
 to admit the hypothesis that specific diseases may be conveyed 
 
 ■ Austin Flint, M.D. Reports and Papers Am. Public Health Assoc, i, pp. 164- 
 172. 
 
24 WATER SUPPLY. 
 
 by the drinking water, and to guard all sources of domestic and 
 public supply from the possibility of contamination by the dejec- 
 tions of persons sick with zymotic diseases and by excremental 
 matter generally. 
 
 What has hitherto been said has had reference to the pollu- 
 tion of water by excremental matter from persons actually sick 
 with communicable diseases. Any pollution, however, by ex- 
 cremental matter should be guarded against, partly because at 
 some time the discharges of sick persons may accompany the 
 other excremental matter, and partly because there is evidence 
 that under certain circumstances the discharges of healthy per- 
 sons and animal matter generally may give rise to disease. Under 
 what circumstances excremental matter becomes dangerous we 
 do not know. It is certain that it is often taken into the stom- 
 achs of men and animals, apparently without doing any harm. 
 Fish often gather about the mouths of sewers and seem to thrive ; 
 the Norwegians save up in summer the dung of the horses and 
 sheep to serve as food for the cattle in winter ; and in many 
 localities where wood is scarce, dried animal excrement is used 
 for fuel and in direct contact with the food to be cooked. Dr. 
 Emmerich, who does not believe in. the " drinking-water theory," 
 himself drank dilute sewage for a number of days in succession, 
 and persuaded several of his patients to do likewise, with no ill 
 effect : * moreover, the experience of every water-analyst shows 
 that there are many grossly polluted wells which have never been 
 known to produce disease. It is thus difficult to believe that in 
 ordinary excremental matter there is any specific poison ; but as 
 there seems to be abundance of circumstantial evidence to show 
 that disease does, at times, follow the use of water which has 
 received excremental pollution, we are forced to believe either 
 that the organic matter undergoes change, and in certain stages 
 of decomposition can produce sickness, or that it is accompanied 
 by something, not yet isolated, which is the real morbific princi- 
 ple, or, indeed, that its effect is to predispose to disease, which 
 fails to attack others not thus predisposed. 
 
 Next in importance to excremental matters is the refuse from 
 slaughter-houses, wool-pulling establishments, tanneries, etc., — 
 animal refuse, in fact, from various sources. Here we are met 
 
 * Zeitschrift ftir Biologic, xiv (1878), p. 591. 
 
DRINKING WATER AND DISEASE. 2$ 
 
 by the objection that much animal matter is consumed in a more 
 or less decayed condition with apparent harmlessness, as for 
 instance, ripe game or ripe cheese : but, to offset this, there are 
 numerous cases on record where sickness has arisen from spoiled 
 meat, especially in the form of sausages and potted meats. But, 
 after all, as stated above, the evidence which connects disease 
 with polluted water is purely circumstantial, and the amount of 
 organic matter, even in waters classed as dangerous, is so small, 
 that, as pointed out by Professor Mallet,* it furnishes " important 
 evidence against any chemical theory of the production of dis- 
 ease from this source, any theory based on the simple assumption 
 that some of the chemical products of the decomposition of 
 organic matter are poisonous or noxious in their effect upon the 
 human system." Instancing two particular waters, he says : " If 
 ike whole of the organic carbon and nitrogen found in such waters 
 as Nos. 35 and 36, of the highly dangerous character of which 
 there can scarcely be a doubt, existed as strychnine, it would be 
 necessary to drink about a half a gallon of the water at once in 
 order to swallow an average medicinal dose of the alkaloid. It 
 is not easy to believe that the ptomaines, or any other chemical 
 products of the putrefactive change as yet observed, can possess 
 an intensity of toxic power so very much greater than that of 
 the most energetic of recognized poisons. While numerous facts 
 go to support the belief that, not to the effect of any chemical 
 substances (such effect necessarily standing in definite relation to 
 their quantity), but to the presence of living organisms, with their 
 power of practically unlimited self-multiplication, we must in all 
 probability look for an explanation of most, at any rate, of the 
 mischief attributable to drinking water, it is of course possible 
 that indirectly a large amount of organic matter in water may be 
 more dangerous than a smaller quantity, as furnishing on a greater 
 scale the suitable material and conditions for the development of 
 noxious as well as harmless organisms." 
 
 Although there are many substances of vegetable origin which 
 are violent poisons, such as the vegetable alkaloids, for example, 
 it is generally held that refuse of vegetable origin is of much less 
 importance as a source of pollution than that coming from animal 
 sources. This is probably true, in general, but it is well known 
 
 * National Board of Health Bulletin, May 27, 1882. 
 
26 WATER SUPPLY. 
 
 that the vegetable refuse from certain manufacturing operations 
 may be very offensive : such, for instance, is the refuse from 
 starch factories, the water in which flax has been retted, etc. 
 That such water would be unfit to drink, unless enormously 
 diluted, one can hardly doubt. To quote again from Professor 
 Mallet : 
 
 " If the theory be accepted, which has so much in its favor, 
 attributing the production of disease by organic matter in drink- 
 ing water not to any specifically poisonous substance or substances, 
 but to the presence and action of living organisms, it seems quite 
 conceivable that a water containing organic matter of any kind, 
 including vegetable matter, may be harmless at one time, and - 
 harmful at another, when perhaps a different stage of fermenta- 
 tion or putrefactive change may have been entered upon, and 
 special organisms may have made their appearance or entered 
 upon a new phase of existence. Thus, there might possibly be 
 safety in drinking a peaty water, or water filtered through beds 
 of dead forest leaves, when fresh ; danger when, after a certain 
 amount of atmospheric exposure, bacterial organisms had become 
 developed ; and safety again, perhaps, after the growth of such 
 organisms had fallen off, and more or less of the available organic 
 matter had been consumed." 
 
 However views may differ as to the possible injury from this 
 or that particular form of contamination, we are safe in accepting 
 the two following principles as fundamental guides in the selec- 
 tion of a water for water supply : 
 
 1. A water suitable for domestic supply must be free from all 
 substances which are known to produce an injurious effect on the 
 human system, or which are suspected with good reason, or on 
 good authority, to produce such an effect. 
 
 2. The water should be, as far as practicable, free from all 
 substances and from all associations which offend the general 
 aesthetic sense of the community, and thus affect the system 
 through the imagination, even if there is good reason to suppose 
 that it is in itself perfectly harmless. 
 
 The first of these principles needs no argument to justify it ; 
 with reference to the second, a word or two may be said. Dr. 
 Emmerich, even, admits that not every one could drink dilute 
 sewage as he did, because the mere idea of so doing, or the sight 
 of a floating hair or other unattractive object, would no doubt 
 
DRINKING WATER AND DISEASE. 2J 
 
 with many persons produce disgust and nausea, and, if the water 
 were forced down, it would most Hkely be thrown up again. 
 While there is no doubt of this power of the imagination and its 
 effect on the physical system, common sense must fix a limit to 
 the apphcation of this second principle, and some latitude must 
 be allowed according to the circumstances of the particular case. 
 Thus, if a lake is chosen as the most available source of supply, 
 the fact that some persons bathe in its waters can hardly condemn 
 it for use ; and the fact that a limited amount of town drainage 
 finds its way into a stream does not necessarily prevent its being 
 used at some portion of its course, although a stream once seri- 
 ously polluted should not be looked upon as an available source 
 of supply. Again, most persons naturally object to v/ater as 
 muddy as that of many of our Western streams, in spite of the 
 favorable testimony of those in the habit of. using it, but by a 
 short residence in St. Louis, for instance, most persons soon be- 
 come accustomed to the turbidity. The turbidity is a real objec- 
 tion to the water, but, in the case of a water like that of the 
 Missouri, a town would not be justified in postponing the intro- 
 duction of the water because it was not able at the same time to 
 adopt a scheme for its thorough filtration. In the same way, if 
 the only objection to a river or pond water is a yellow or brown- 
 ish-yellow color derived from vegetable, especially peaty matter, 
 the water need not be condemned, although most persons would 
 prefer a colorless water. 
 
 Undoubtedly, the best water for drinking is a moderately soft 
 spring water, in which all possibility of contamination is out of 
 the question. Unfortunately, however, it is comparatively sel- 
 dom that such water is available in quantities sufficient for the 
 supply of large towns. Many spring waters are so hard that, 
 while not unsuited for drinking, they are unsuited for many man- 
 ufacturing uses, for use in steam boilers, for washing and culinary 
 purposes. It is a mistake to claim that the water which is abso- 
 lutely best for drinking must be chosen, at any expense, as a town 
 supply; when a soft surface water, free from appreciable pollution 
 can be obtained, it entails a very serious and constant expense to 
 reject it in favor of a hard water, which may, to be sure, be clearer 
 to the eye and somewhat more pleasant to the taste. There are 
 surface waters and there are surface-water supplies which are un- 
 doubtedly bad, but a good surface water, such as may be taken 
 
28 WATER SUPPLY. 
 
 directly from many streams or such as may be obtained from deep 
 lakes and from proper storage basins, is perfectly well suited for 
 domestic use or for town supply. There are some who maintain 
 an opinion contrary to that which has been expressed. The 
 Vienna Commission in 1864, rejected surface waters from among 
 the waters suitable for domestic use, on the ground of their vari- 
 able temperature (see page 91) and their liability to pollution. 
 The German Pubhc Health Association, at the Danzig meeting 
 in 1 880, by a small majority and after a lively discussion, adopted 
 a resolution to the effect that spring water or properly protected 
 ground water were the only admissible sources of supply ; but two 
 years later this dictum was modified so as to include filtered river 
 water as fulfilling the required conditions, and this conclusion is 
 sanctioned by practice and experience. 
 
CHAPTER II. 
 
 WATER ANALYSIS. 
 
 It is not the purpose of the present chapter to give in detail 
 all the various methods employed in the analysis of water, or to 
 serve as a guide to those wishing to make such analyses. Ex- 
 cellent manuals on this subject already exist. An attempt will 
 be made, however, to explain what is the meaning of the various 
 terms used in the reports of water analysis, and to indicate the 
 significance to be attached to the figures given. 
 
 For this discussion we may classify the substances occurring 
 in a natural water according to the following scheme : 
 
 1. Suspended matter ] ^"^S'" i Animal. 
 
 1 Vegetable. 
 
 2. Dissolved matter j Gaseous. 
 
 10rgamc....jAmmal.^^^ 
 
 Suspended matter. — The determination of the amount of sus- 
 pended matter is principally of value with reference to possible 
 schemes of sedimentation or filtration. The operation is com- 
 monly conducted by passing a measured quantity of water through 
 a paper filter which has previously been dried at ioo° C. and then 
 weighed ; after the passage of the water, the filter with its con- 
 tents is again dried at ioo° and weighed for the second time. 
 The difference between the two weights is the weight of the 
 suspended matter. The filter with its contents is then trans- 
 ferred to a platinum crucible, and the crucible is heated until the 
 filter has been burned up and the organic matter destroyed. 
 The weight of what remains, minus the weight of the ash which 
 a filter, such as was used, is known to leave, is the weight of the 
 inorganic or mineral portion of the suspended matter. Another 
 method, which is exact enough for most purposes, consists in 
 evaporating a measured quantity of the water in its natural con- 
 dition, and an equal quantity which has been filtered through 
 paper. The residues in the two cases are weighed and the dif- 
 ference is approximately the weight of the matter which was in 
 suspension. 
 
30 WATER SUPPLY. 
 
 It is quite as important to know the character of the suspended 
 matter as to know its absolute amount. It is desirable to know 
 whether it settles readily, and, if it contains much organic matter, 
 whether this is mainly of animal or of vegetable origin. Some 
 information as to the latter point may be obtained by observing 
 the appearance and odor when it is strongly heated, but micro- 
 scopical examination is of the greatest service in this connection. 
 
 In expressing in figures the results of the analysis, a consid- 
 erable difference exists in the practice of different chemists as to 
 the unit to be employed. The methods of expression in most 
 common use are the following: 
 
 (i) In grains to the English (imperial) gallon, which meas- 
 ures 277 cubic inches and is equivalent to lo lbs., or 70,000 grains, 
 of pure water. This method is still quite common in England. 
 
 (2) In grains to the U. S. gallon, which measures 231 cubic 
 inches and is equivalent to 58,372 + grains of pufe water. This 
 method is very common in the United States. 
 
 (3) On a decimal basis, as so many parts (by weight) in 1,000, 
 10,000, 100,000, or 1,000,000 parts of the water according to the 
 amount of dissolved matter present. Thus, mineral waters are 
 usually reported as containing so many parts in 1,000 or in 10,000 
 parts, while for potable waters parts in 100,000 or 1,000,000 are 
 employed. The preference of the author is decidedly for " parts 
 in 100,000," which is now generally used in France and Germany; 
 also in the Reports of the Rivers Pollution Commission of Great 
 Britain, and, in this country, in the reports of the National and 
 of many State Boards of Health. 
 
 (4) As so many milligrams to the liter. This.would be equiv- 
 alent to so many parts in 1,000,000 if the water possessed the 
 same density as pure water ; that is, if a liter of the water actually 
 weighed 1,000 grams. Practically, the error introduced by meas- 
 uring instead of weighing the water taken for analysis, is, in the 
 case of most potable waters, less than the error of analysis, thus: 
 
 If the specific grav- 
 ity is 
 
 1. 010 
 
 1. 010 
 
 1.020 
 
 1 . 020* 
 
 1.040 
 
 1.040 
 
 * The specific gravity of sea water is about i.02g. 
 
 So many milligrams 
 to the liter 
 
 Equal so many parts 
 in 1,000,000. 
 
 DifE. 
 
 5.000 
 
 4-950 
 
 0.050 
 
 10.000 
 
 9.901 
 
 0.099 
 
 5.000 
 
 4.902 
 
 0.098 
 
 10.000 
 
 9.804 
 
 ig6 
 
 5.000 
 
 4.808 
 
 0.192 
 
 10.000 
 
 9.615 
 
 0.385 
 
WATER ANALYSIS. 3 1 
 
 In the case of mineral waters, sea-water, etc., the difference 
 is too great to be neglected ; and, in such cases, it should always 
 be distinctly stated whether the results are in milligrams to the 
 liter, or in parts in 1,000,000 by weight. 
 
 Dissolved substances. — Gases. — Almost all gases dissolve in 
 water to a greater or less extent. It is seldom, however, that a 
 water which is' proposed or used as a source of supply contains 
 appreciable quantities of any gases other than those of the 
 atmosphere — oxygen, nitrogen, and carbonic acid. These gases 
 are present in very varying proportions. Oxygen is found in 
 all waters which have been exposed to the air. The waters of 
 artesian wells often contain none of this gas, but they absorb it 
 at once on reaching the air ; waters from highly polluted sources 
 are also deficient in oxygen. Nitrogen occurs to a greater or 
 less extent in all waters, and in some mineral or effervescent 
 waters.it forms the largest part of the dissolved gases. Carbonic 
 acid is present in all potable waters ; its presence is of important 
 influence in determining the amount of carbonate of lime which 
 a given water can contain, and the solvent effect of water on 
 many minerals is due to the presence of carbonic acid. 
 
 It was formerly the general custom to make a complete quan- 
 titative analysis of the various gases present in a sample of water 
 under examination. The analysis was made by boiling out the 
 gases from a measured quantity of water and submitting the 
 mixture obtained to the accurate but tedious methods of gas 
 analysis. When, however, as often happened, the determinations 
 were not made on the spot, but, on the contrary, on samples of 
 water which had been standing for days or for weeks, they were 
 nearly worthless, and are now rarely made except in the case of 
 mineral waters. In fact, at the present time, it is seldom that 
 the amount of any dissolved gas is determined, except oxygen ; 
 and for determining the amount of oxygen in a given water, a 
 process was devised a few years ago by Schutzenberger which 
 admits of being performed with sufficient accuracy out of doors. 
 A water which is polluted by decaying organic matter not only 
 calls upon the air about it to furnish oxygen for the combustion 
 of the decaying matter, but uses up in the process more or less, 
 sometimes all, of the oxygen previously dissolved in the water. 
 The amount of dissolved oxygen is thus considered by some to 
 be an indication, in the inverse direction of course, of the 
 
32 WATER SUPPLY. 
 
 amount of impurity present. An example of the application of 
 this method may be found on page6i. The results thus ob- 
 tained are of value, as in this case, in tracing the course of the 
 same polluted stream, but a knowledge of the absolute amount 
 of dissolved oxygen gives no means of judging of the purity of a 
 single sample of water ; for, not only is the water of some arte- 
 sian wells free from oxygen, as stated above, but the ground 
 water generally and the water of unpolluted springs and deep 
 wells is also deficient in oxygen. 
 
 In the case of all the gases, the amount present is indicated 
 as so many cubic inches to the gallon, or as so many cubic centi- 
 meters to the liter. 
 
 Total solids in solution. — The determination of the total 
 amount of dissolved matter is made by evaporating a measured 
 quantity of water to dryness in a weighed platinum dish, drying 
 at some definite temperature, and then weighing the dish with its 
 contents. The determination is, at the best, a rough one, and 
 the solid matter obtained by evaporation does not exactly repre- 
 sent what was originally in solution. In the evaporation some 
 substances pass off with the steam and are lost. Other substan- 
 ces are changed in character by the treatment. If the residue be 
 dried at the temperature of boiling water, which is that most com- 
 monly employed, some of the salts retain water of crystalliza- 
 tion ; at a somewhat higher temperature, even as low as 140° C, 
 the organic substances begin to be decomposed and lose weight. 
 In spite of this, some chemists use a temperature as high as 180° C. 
 It is generally of no importance for sanitary purposes that the de- 
 termination should be exact, for it is of no consequence whether 
 a water leaves 4 or 6 parts of " total solid residue," whether it 
 leaves 10 or 12 parts, but in the case of periodical examinations 
 of the same water the determinations should be made in the same 
 way at all times, in order that they may be compared with each 
 other. 
 
 It is seldom necessary for sanitary purposes to make a com- 
 plete analysis of the water and to determine the amount of each 
 of the various compounds which make up the "total solids." A 
 few of the mineral constituents are almost always determined for 
 special reasons, and, in individual cases, others may be estimated, 
 but, as a rule, the indications of qualitative tests sufifice. In a 
 water which has passed through metal pipes, or which receives 
 
WATER ANALYSIS. 33 
 
 the flow from mines, or from manufacturing operations, poison- 
 ous metals should be looked for ; lead and copper are the most 
 common, and less than one-tenth of a grain of lead to the gallon 
 has been known to do harm ; arsenic is frequently found in run- 
 ning streams and sometimes in water which contains, or rather 
 which deposits, oxide of iron, but the amount is usually too small 
 to notice ; arsenic has, however, been the cause of well-pollution 
 in the neighborhood of manufactories of aniline colors. 
 
 Chlorine. — Chlorine is almost always determined quantitatively. 
 It will be understood that this element never occurs free in nat- 
 ural waters, but always in combination as chloride of sodium or as 
 some other chloride. The amount is generally reported as so much 
 chlorine, although some analysts prefer to calculate the corre- 
 sponding amount of chloride of sodium (common salt), and report 
 in this way. Although chlorides are present in all soils and nat- 
 ural waters, the quantity in the uncontaminated water of most 
 regions is very small. Where this is known to be the case, the 
 presence of any noticeable amount of chlorine (say much more 
 than I part in 100,000) indicates contamination from human 
 sources, as chloride of sodium is a constant constituent of sewage 
 and of animal refuse in general, and is not eliminated to any con- 
 siderable extent either in passing through the soil or by the action 
 of vegetation. Chlorides remain as evidences of past contami- 
 nation after all other evidences have ceased to exist, although 
 the amount present does not bear any constant proportion to the 
 amount of polluting substances which are or have been in the 
 water. 
 
 Hardness. — A determination of the " hardness " usually finds 
 a place in the examination of % water proposed as a source of 
 supply. The determination is made by taking advantage of one 
 of the properties which make hard water undesirable — namely, the 
 property of destroying soap. A solution of soap is prepared of 
 such a strength that a measured quantity of it is exactly destroyed 
 or neutralized by a known amount of some compound of lime. 
 The lime compound is previously dissolved in a certain amount, 
 say 100. cubic centimeters, of water, and the test is made by ascer- 
 taining how much of this same standard soap solution is destroyed 
 by 100 c.c. of this particular water. The hardness of another 
 portion of the water is determined after boihng for some time ; 
 this is called ^^ permanent hardness" and is due to sulphates 
 3 
 
34 WATER SUPPLY. 
 
 and other soluble compounds of lime and magnesia. The total 
 hardness, less the permanent hardness, is the " temporary hardness^' 
 and is due to the bicarbonates of lime and magnesia, which are 
 decomposed by boiling. The hardness is generally expressed in 
 degrees, which have different significations in different countries. 
 In England, where the process originated, a degree of hardness 
 corresponds to a grain of carbonate of lime in one imperial gallon 
 of water ; for example, a water of 5 degrees hardness means a 
 water, each gallon of which contains compounds of lime or mag- 
 nesia or both equivalent in soap-destroying power to 5 grains of 
 carbonate of lime. In Germany the degrees of hardness indicate 
 the equivalent of so many parts of oxide of calcium (quicklime) 
 in 100,000 parts of water. In France the degrees mean so many 
 parts of carbonate of lime in 100,000 parts of water. In America, 
 in spite of the anomaly, many express the hardness in English 
 degrees, i. e., in grains to the imperial gallon, while the other re- 
 sults are given in grains to the United States gallon. The 
 French system of parts in 100,000 is, however, to be preferred. 
 
 Combined nitrogen. — Nitrogen exists in potable waters under 
 a variety of forms of combination. Animal matter generally con- 
 tains nitrogen, as do also many substances of vegetable origin: 
 this is usually spoken of as " organic nitrogen." Nitrogen also 
 exists in the form of ammonia and ammoniacal salts, and these 
 compounds are due almost entirely to the decomposition of nitrog- 
 enous vegetable and more especially animal matters. For this 
 reason, although in itself harmless, the ammonia is determined 
 quantitatively. The amount present, even in a polluted water, is 
 small, and when expressed in figures — as in the various tables on 
 following pages — seems insignificant, nevertheless, as a sign of or- , 
 ganic pollution, its determination should not be omitted. Although 
 the amount of ammonia is so small — a water containing o.i part 
 by weight in 100,000 parts of the water being grossly polluted — 
 the figures are entitled to confidence because the method usually 
 employed is very delicate, and, by a process of distillation, the 
 ammonia in a large volume of water may be concentrated into a 
 small bulk. The " Nessler test," which is employed in estimating 
 ammonia, is capable of detecting i part by weight of ammonia in 
 20,000,000 parts of water. 
 
 A great diversity of opinion exists as to the value which at- 
 taches to a determination of the exact amount of nitrogen pres- 
 
WATER ANALYSIS. 35 
 
 ent as nitrites and nitrates. The compounds themselves, in small 
 amounts, are no doubt harmless, but it is also true that they re- 
 sult from the oxidation of nitrogenous organic matter. The 
 nitrifying process is now believed to take place under the in- 
 fluence of minute organisms, and the process goes on in the vari- 
 ous soils at very different rates according to the alkalinity of the 
 soil, to the amount of moisture and to other conditions. More- 
 over, the nitrates once formed may again be reduced to ammoni- 
 acal compounds, or even to free nitrogen, so that the nitrates 
 cannot be a quantitative indication of the amount of pollution. 
 
 In spite of this, Frankland lays great stress on the exact deter- 
 mination of nitrogen in this form, and has introduced into water 
 reports the term " previous sewage contamination." The figures 
 given, in any case, under this head are reached by determining 
 the total amount of nitrogen which is present as ammonia, and 
 also as nitrites and nitrates ; after subtracting the small amount 
 of nitrogen which rain water contains in these forms, there is 
 calculated from the remainder how much of what is called " aver- 
 age London sewage " would be necessary to account for this 
 amount of nitrogen. The composition of the so-called " average 
 London sewage" is not, as might be supposed, deduced from 
 a considerable number of examinations made at various times, 
 but the average sewage is taken arbitrarily as containing lo parts 
 of combined nitrogen in 100,000 parts. 
 
 Mallet (in a report already cited) is " inclined to attach spe- 
 cial and very great importance to a careful determination of the 
 nitrites and nitrates in water to be used for drinking." 
 
 It is true that in consecutive examinations of the same water 
 it is satisfactory to know exactly the amount of nitrates and the 
 variation from time to time ; but, in case of a single examination, 
 it is generally sufficient to know whether there is much or little, 
 or none. The test most commonly applied, although not the 
 most delicate, is the sulphate of iron (ferrous sulphate) test. A 
 small quantity of the water under examination is mixed in a 
 glass tube with an equal volume of pure concentrated sulphuric 
 acid. The mixture, which becomes very hot, is cooled to the 
 temperature of the air, and there is then poured upon it a solu- 
 tion of sulphate of iron. If nitrates are present, a dark ring or 
 layer forms between the two liquids. The amount of nitrates 
 present may be inferred from the extent to which the water 
 
36 WATER SUPPLY. 
 
 must be concentrated before it will give indications by this test, 
 but the indications are strictly comparable only when the test is 
 performed in precisely the same way and by the same person. 
 
 It should be understood that the argument drawn from the 
 presence of nitrates is this : The nitrates indicate the previous 
 existence of nitrogenous organic matter, which has been oxidized 
 and converted into harmless compounds by natural agencies. 
 If these same agencies could be relied on indefinitely it would be 
 well, but no one can tell at what moment — owing to increase in 
 the amount of the polluting substances, or to their gradual accu- 
 mula,tion in the soil, or to other changed conditions — the natural 
 agencies may prove inadequate to the task and allow incom- 
 pletely oxidized and harmful substances to reach the source of 
 supply. 
 
 Organic matter. — Of the polluting material which reaches 
 water which may be used for drinking, the organic portion is felt 
 to be that which directly or indirectly introduces the element of 
 danger. Just how and from what particular substances the dan- 
 ger arises is unknown, and it is extremely doubtful whether the 
 dangerous something will ever be an object of chemical deter- 
 mination, but in our present ignorance it is, generally felt that 
 it is desirable to obtain such indications as are possible of the 
 amount and character of the organic matter present. 
 
 To a person unfamiliar with chemistry, it might seem to be 
 no difficult task to determine exactly how much matter of animal 
 and how much matter of vegetable origin is present in a given 
 water. The truth is, however, that it is not only difficult but 
 impossible, either to determine the total amount of organic mat- 
 ter or to decide upon its origin. As far as the total amount is 
 concerned, the fact that it cannot be determined is a matter of 
 no great consequence. Even if the chemist could say with cer- 
 tainty that a particular water contained exactly i or 2 or 5 parts 
 of organic matter in 100,000, we should be far from having the 
 data necessary to form an opinion as to the wholesomeness of the 
 water, for it is evident to any one that a pound of sugar or of 
 glycerine would have a very different importance from that of a 
 pound of (dry) faeces, yet either is a pound of organic matter. 
 
 Formerly it was the general custom to subject the residue of 
 evaporation (" total dissolved solids ") to the action of a low red 
 heat until all the carbonaceous matter was destroyed and to deter- 
 
WATER ANALYSIS. 37 
 
 mine the loss of weight. Thus was obtained what is now gen- 
 erally tabulated as " loss on ignition," or "organic and volatile 
 matter ; " but the loss of weight is far from being entirely due to 
 the destruction of organic matter. According to the degree 
 of heat applied and the length of time during which it is con- 
 tinued, there is more or less loss due to the volatihzation of 
 alkaline chlorides. There is also loss from the decomposition 
 and partial volatilization of several compounds ; thus, the car- 
 bonates of lime and magnesia are more or less completely con- 
 verted into oxides by expulsion of the carbonic acid, or in the 
 presence df a sufficient quantity of silicic acid, into silicate?. The 
 nitrates are converted into carbonates, oxides, or silicates, chlo- 
 ride of magnesium is decomposed in the presence of hydrated 
 compounds with escape of chlorhydric acid, and other changes 
 take place which it is not necessary to particularize. 
 
 For these reasons, no two persons are likely to obtain the 
 same result from the same water, and not much value is attached 
 nowadays to the determination ; it is sometimes of assistance in 
 forming the final opinion in case of a doubtful water, but the way 
 in which the residue acts when heated gives more information 
 ■ than a knowledge of the loss of weight. 
 
 Of the more modern methods in somewhat general use for 
 reaching information about the organic matter in water, there 
 are three which may be mentioned here, and which will be spoken 
 of as the " permanganate " method, the " ammonia " process, and 
 " Frankland's " method.* All of the processes possess a certain 
 value, and all are widely open to criticism. 
 
 Permanganate of potash is a highly colored crystalline salt 
 soluble in water, to which, even if the solution be very dilute, it 
 communicates a marked pink color. This compound, which 
 contains a considerable proportion of oxygen, possesses the 
 property of oxidizing, with more or less readiness, most forms of 
 organic matter, being itself destroyed in the process and losing 
 the characteristic color. » By successive additions of a perman- 
 
 * These various processes will be discussed only briefly in this place. They 
 have recently been thoroughly investigated by Professor J. W. Mallet, under direc- 
 tion of the National Board of Health. The full report has not yet appeared, but a 
 preliminary report was published as a Supplement (No. 19) to the National Board of 
 Health Bulletin, May 27, :882. 
 
38 WATER SUPPLY. 
 
 ganate solution of known strength until the color persists, it 
 is possible to determine how much permanganate is destroyed 
 by a known volume of a given water. There are various ways 
 of applying the permanganate solution. Some prefer to use the 
 reagent in alkaline, and others in acid solutions ; some heat the 
 liquid to one temperature and some to another.* The results 
 obtained are reported by stating how many parts, by weight, of 
 the crystallized permanganate are required for 100,000 parts by 
 weight of water ; or how many parts, by weight, of oxygen (from 
 the permanganate) are used up in the process. Some, indeed, 
 assume an arbitrary number, by which they multiply the amount 
 of permanganate employed, and call the result organic matter. 
 Where the expression " organic matter " occurs in German re- 
 ports, the figures are probably obtained by multiplying the 
 amount of oxygen used up by 20,f but in American reports the 
 expression is quite likely to mean simply the " loss on ignition." 
 The results of the permanganate method cannot have an absolute 
 value, because different organic substances vary in the complete- 
 ness with which they are oxidized under the same conditions; 
 and, even if they were all completely oxidized under certain 
 attainable conditions, it would still be true that one gram of one 
 kind of organic matter would require a very different amount of 
 oxygen from that which would be required by one gram of some 
 other kind of organic matter. Moreover, with the same water, 
 the results differ very much according to the method of employ- 
 ing the test, so that to be able to give a useful interpretation to 
 any particular results, it is necessary to know what method was 
 followed by the analyst. 
 
 The so-called ammonia method of water analysis was devised 
 by the English chemists Chapman, Wanklyn and Smith, J and 
 has been much used in England and in this country. In this 
 
 * See Kubel's Anleitung zur Untersuchung von Wasser, bearbeitet von Dr. Ferd. 
 Tiemann, Braunschweig, 1874. Also a. paper by Dr. Tidy, Chemical News, xxxvii 
 (1878), p. 283. 
 
 •)• Or by multiplying the amount of permanganate consumed by 5. This is accord- 
 ing to Dr. Woods, but Dr. Letheby prefers to multiply the amount of permanganate 
 by 8. Whatever number be used, it frequently happens that the " organic matter,'' 
 as thus estimated, exceeds the weight of the " total dissolved solids." 
 
 X Water Analysis. By J. A. Wanklyn and E. T. Chapman. 5th edition, 
 rewritten by J. A. Wanklyn, London, 1879. 
 
WATER ANALYSIS. 39 
 
 method, advantage is taken of the fact that certain kinds of ni- 
 trogenous organic matter, when treated with a strongly alkaline 
 solution of permanganate of potash, give off a definite portion or 
 the whole of their nitrogen as ammonia ; and the value of the 
 method lies in the assumption that it is the nitrogenized organic 
 matter which is to be regarded as the chief source of danger in 
 polluted water. In working the ammonia method, the water 
 under examination is put into a retort, tnade alkaline by means 
 of carbonate of soda, and distilled as long as the water which 
 condenses contains enough ammonia to be measured by Nessler's 
 solution. Then a solution of caustic soda and permanganate of 
 potash is added, and distillation is continued. Another portion 
 of ammonia now comes off, owing to the action of the perman- 
 ganate on the nitrogenous organic matter. The amount of 
 ammonia thus obtained is determined, and is tabulated as "al- 
 buminoid ammonia," because albumin is one of the bodies which 
 acts in this way. 
 
 It will thus be understood that the so-called "albuminoid 
 ammonia" is not something which exists in the water ready 
 formed ; moreover, because different amounts are obtained from 
 the same weight of different nitrogenous substances, as well as 
 on account of the fact already alluded to, that the oxidation of 
 different substances by the permanganate of potash is more or 
 less incomplete, the figures obtained have a relative rather than 
 an absolute significance. 
 
 Although this method does not accomplish all that might be 
 des'ired, the results, properly interpreted, are of great value, and 
 the method has been of immense service in the cause of public 
 heahh. 
 
 The third method, known as the "combustion " process, or 
 Frankland's method, was devised by Frankland and Armstrong, 
 and used by the Rivers Pollution Commission of Great Britain 
 in the examination of the very large number of waters, the anal- 
 yses of which appear in the various reports of the Commission. 
 This method is by far the most elaborate of any that have been 
 proposed: it consists in evaporating a given quantity of the 
 water, under carefully regulated conditions, and in submitting 
 the residue to a process of organic analysis, by which all the 
 carbon is converted into carbonic acid and the nitrogen is liber- 
 ated in the gaseous state. The mixture of nitrogen and carbonic 
 
40 WATER SUPPLY. 
 
 acid is then analyzed by processes of gas analysis. The results 
 are stated in so many parts of " organic carbon " and so much 
 " organic nitrogen," in 100,000 parts of the water, and sometimes 
 the sum of the two is spoken of as the amount of the " organic 
 elements." The character of the organic matter is inferred from 
 the relative proportion of carbon and nitrogen. 
 
 The process, as Frankland himself says, is " both troublesome 
 and tedious." The apparatus employed is frangible and some- 
 what costly, and the manipulative skill of a trained chemist is 
 required to carry out the work. It is moreover a process that 
 cannot be taken up off-hand, even by a trained chemist, but one 
 requiring tolerably constant practice. As Mallet says : " It is 
 better adapted to regular use in the examination of many 
 samples of water in a large public laboratory than to occasional 
 use by a private individual in now and then examining a single 
 water." 
 
 Although the combustion process is the most elaborate, and, 
 in some respects the most satisfactory that we have, even this, 
 "in its present form cannot be considered as 'determining' the 
 carbon and nitrogen of the organic matter in water in a sense to 
 justify the claim of ' absolute ' value for its results which has 
 been denied to those of all other methods. It is but a method 
 of approximation, involving sundry errors, and in part a balance 
 of errors." * 
 
 Standards of purity. — Having followed this brief description 
 of the methods of analysis most frequently employed, a person, 
 who is not a chemist, may naturally ask for some directions in 
 order to interpret the figures reported by the analyst. Now, it 
 is true that it may be possible to fix certain numerical limits 
 and to reject without hesitation all waters exceeding these 
 limits ; but there always will be difficulty in deciding how near 
 to any limit a suspicious water may come and still be used with 
 a reasonable degree of safety. To condemn a water without 
 sufficient cause is, of course, undesirable, as the procuring a dif- 
 ferent supply may involve considerable expense. 
 
 Moreover, it cannot be insisted upon too strongly that differ- 
 ent classes of water cannot be judged by the same standard, and. 
 
 * For a full discussion of the sources of error, see Mallet's report, already alluded 
 to ; .also, Amer. Chem. Journ., iv (1883). 
 
WATER ANALYSIS. 
 
 41 
 
 the results of the analysis of waters belonging to different classes 
 ought not to be put into the same table or otherwise arranged so 
 as to invite comparison. If within the same geological area it is 
 possible to analyze the water from a considerable number of un- 
 polluted wells, a standard may be fixed for the well water of that 
 region, and a surface water may be compared with other surface 
 waters of the same or of a similarly situated region ; or a stream in 
 one part of its course may be compared with its own unpolluted 
 head waters. To fix, however, a definite standard -which will 
 apply to all waters and by which any one can judge of a given 
 water from the numerical results of analysis is impracticable. 
 Every doubtful water must be considered by itself with all the 
 light that can be brought to bear upon it. 
 
 Bearing in mind what has just been said, we may note the in- 
 terpretation which is given to the results of the several methods 
 in common use for determining " organic matter," or, at least, 
 for obtaining indications as to its amount and character. Wan- 
 klyn's own interpretation of the results of the ammonia process 
 is essentially as follows : 
 
 If a water yield no " albuminoid ammonia," it may be passed 
 as organically pure, despite of much free ammonia and chlorides ; 
 a water giving less than 0.005 part of " albuminoid ammonia " 
 in 100,000 parts may be regarded as very pure. A water contain- 
 ing 0.005 part of " albuminoid ammonia " together with a con- 
 siderable quantity of free ammonia is suspicious, but in the 
 absence of free ammonia, the "albuminoid ammonia" maybe 
 allowed to amount to something like o.oio part ; above o.oio 
 should be regarded as very suspicious, and according to Wanklyn 
 over 0.015 part should condemn the water. 
 
 The Rivers Pollution Commission, in interpreting the results 
 of Frankland's process,- make the following classification : " Sur- 
 face water or river water which contains in 100,000 parts more 
 than 0.2 part of organic carbon or 0.03 part of organic nitrogen 
 is not desirable for domestic supply, and ought, whenever practi- 
 cable, to be rejected. Spring and deep-well water ought not to 
 contain in 100,000 parts more than o.i part of organic carbon or 
 0.03 part of organic nitrogen." 
 
 Dr. Frankland, while " deprecating a hard and fast division 
 of -yater into classes," suggests a rough classification according 
 to the amount of organic carbon present; this classification is 
 
42 
 
 WATER SUPPLY. 
 
 given in Table III, the figures indicating such a fraction of one 
 part by weight of organic carbon in ioo,ooo parts by weight of 
 the water. 
 
 Dr. Tidy, while admitting the impossibihty of deciding of the 
 quality of a water from an incomplete analysis suggests certain 
 limits as a guide to the interpretation of the results obtained by 
 the permanganate process, conducted, be it understood, accord- 
 ing to a particular method. These also appear in Table III. 
 
 TABLE III.— Classification of Natural Waters. 
 Parts in 100,000. 
 
 Classification. 
 
 Organic ( 
 Dr. Fra 
 
 Upland sur- 
 face water. 
 
 Carbon. — 
 
 NKLAND. 
 
 Other 
 
 waters. 
 
 OXYGEN REQUIRED TO OXI- 
 DIZE Organic Matter.— 
 De. Tidy. 
 
 I Water of great organic purity . 
 
 II Water of medium purity 
 
 Ill Water of doubtful purity 
 
 IV Impure water 
 
 0. -0.2 
 0.2-0.4 
 . 4-0 . 6 
 0.6 + 
 
 O.-O.I 
 
 0.1-0.2 
 0.2-0.4 
 0.4 + 
 
 0. -0.05 
 0.05-0.15 
 0.15-0. 21 
 0.21 + 
 
 Some of the limiting amounts which have been suggested by 
 other chemists are given in Table IV : 
 
 TABLE IV. — Standards of Purity. 
 Parts in ioc,ooc. 
 
 Reichardt 
 
 Kubel and Tiemann 
 
 Wibel 
 
 Fischer 
 
 Total 
 Solids. 
 
 Organic 
 
 Matter. 
 
 Nitric Acid 
 (N.O^). 
 
 Chlorine. 
 
 50 
 
 2. 
 
 0.4 
 
 0.2-0.8 
 
 50 
 
 5- 
 
 0.5-1.5 
 
 2-3 
 
 50 
 
 5- 
 
 0.5-2. 
 
 3-5 
 
 50 
 
 4- 
 
 2.7 
 
 3-5 
 
 Total 
 Hardness. 
 
 18 
 
 18-20 
 
 l8-20 
 
 17 
 
 Professor Mallet, in the report already alluded to, after record- 
 ing the results of an examination into the various processes of 
 analysis which have been briefly described, draws certain general 
 conclusions as follows : 
 
 "I. It is not possible to decide absolutely upon the whole- 
 someness or unwholesomeness of a drinking water by the mere 
 use of any of the processes examined for the estimation of 
 organic matter, or its constituents. 
 
 " 2. I would even go further, and say that, in judging the 
 sanitary character of a water, not or^ly must such processes be 
 used in connection with the investigation of other evidence of a 
 
WATER ANALYSIS. 43 
 
 more general sort, as to the source and history of the water, but 
 should even be deemed of secondary importance in weighing the 
 reasons for accepting or rejecting a water not manifestly unfit 
 for drinking on other grounds. 
 
 " 3. There are no sound grounds on which to establish such 
 general ' standards of purity ' as have been proposed, looking 
 to exact amounts of- organic carbon or nitrogen, ' albuminoid 
 ammonia,' oxygen of permanganate consumed, etc., as perrriis- 
 sible or not. Distinctions drawn by the application of such 
 standards are arbitrary, and may be misleading. 
 
 "4. Two entirely legitimate directions seem to be open for 
 the useful examination by chemical means of the organic con- 
 stituents of drinking water, namely, first, the detection of very 
 gross pollution, such as the contamination of the water of a well 
 by accidental bursting or crushing of soil-pipes, extensive leakage 
 of drains, etc., and secondly, the periodical examination of a 
 water supply, as of a great city, in order that, the normal or 
 usual character of the water having been previously ascertained, 
 any suspicious changes which from time to time may occur shall 
 be promptly detected and their cause investigated. 
 
 " 5. In connection with this latter application of water analy- 
 sis, there seems to be no objection to the establishment of local 
 ' standards of purity ' for drinking water, based on sufficiently 
 thorough examination of the water supply in its usual con- 
 dition." 
 
 These conclusions are given, as they coincide almost exactly 
 with what the author has frequently had occasion to assert, and 
 has for years tried to teach. " In the majority of cases, chemical 
 examination alone cannot be relied upon as giving conclusive 
 evidence as to the suitability of a water for drinking. Of course, 
 if a water is hard, the chemist can say, without hesitation, that 
 the water is unsuited for supply on account of its probable effect 
 on steam boilers, and because it will be uneconomical for use in 
 washing. If the water contains arsenic or lead or other poison- 
 ous metal, the chemist can discover it. If the water is grossly 
 polluted, or is of exceptional purity, chemical examination can 
 determine these facts ; but, in a vast majority of cases, while 
 chemistry may teach something and aid in the decision, it cannot 
 teach everything, and it cannot decide. Now, it would be very 
 convenient, if it were possible, to take each item which is made 
 
44 WATER SUPPLY. 
 
 the object of analytical determination, and say that a good water 
 may contain so much, and if a water contains more, it is not 
 good. This is impossible : a certain amount of the same sub- 
 stance might in one case be a sign of fearful contamination, while 
 in another it might indicate only a normal constituent of the 
 water. 
 
 " In view of the impossibility of saying exactly what is and 
 what is not harmful, any considerable departure from the normal . 
 character of the water in a given locality should be regarded with 
 suspicion. It is true that various students of the matter of water 
 supply have formulated ' standards ' which a water may not 
 overpass. They are, however, only of relative value." * 
 
 It should not be inferred that the chemical analysis is value- 
 less because it cannot furnish data for absolute decision, but it is 
 a great mistake to suppose that the proper way to consult a 
 chemist is to send a sample of water in a sealed vessel with no 
 hint as to its source. On the contrary, the chemist should know 
 as much as possible as to the history and source of the water, and 
 if possible should take the samples himself — that is if he is to ex- 
 press an opinion as to the suitableness of the water for drinking. 
 Nor is it sufficient that some other competent person should 
 possess the knowledge of the locality, etc., and endeavor to in- 
 terpret the figures furnished by the chemist. There are many 
 things which guide the chemist that he cannot put into numer- 
 ical results or even on to paper at all. Many observations are 
 made in the course of an analysis by an experienced person, of 
 which he is himself hardly conscious, but which aid in making up 
 the final opinion. Foxf very truly says : 
 
 " It is a golden rule in water analysis never to give an opinion 
 unless the analyst knows (i) the nature of the source of a water — 
 whether it comes from a spring or well or river or rain reservoir, 
 etc.; (2) the depth of the well, if it is withdrawn from one ; (3) 
 the geology of the district from which it is derived, together with 
 the character of the soil and subsoil ; (4) the distance from the 
 source of the water of the nearest filth or drain." 
 
 At the present time, chemical examination, in connection 
 with a thorough knowledge of a proposed source of supply, must 
 
 * Buck's Hygiene, Vol. i, p. 303. 
 
 f Sanitary Examination of Water, Air and Food. London, 1878. 
 
WATER ANALYSIS. 45 
 
 be the main guide in the selection or rejection of a water ; but 
 there is reason to hope that eventually the decision may be 
 thrown largely upon the biologist. Investigations have been made 
 from a biological stand-point in two directions: (i) by an exam- 
 ination of the organisms in the water itself or which develop in 
 it when the water is allowed to stand ; (2) by injecting the concen- 
 trated water, or a solution of the residue of evaporation, under 
 the skin of rabbits or other animals, and observing the effect on 
 the temperature, etc., of the animals experimented upon. 
 
 With reference to the first method, it may be said that a 
 microscopical examination of the suspended and sedimentary 
 matter should always accompany the chemical examination, and 
 there is no difficulty in recognizing grains of starch, fibers of 
 cotton, silk, wool, etc., and many other sorts of animal and 
 vegetable debris if present ; of course, substances like those men- 
 tioned are evidence of more or less contamination. With regard 
 to living organisms, we do not know as much as we have a right 
 to hope to know about their connection with the wholesomeness 
 of a water. It may be said, in a general way, that bacteria occur- 
 ring in considerable numbers are a sign of impure water, as are 
 also certain infusoria, such as Paramecium, vorticella, monas, etc. 
 The occurrence of leptothrix, crenothrix, etc., are also suspicious 
 signs, but the diatoms and the green algas, as a rule, do not indi- 
 cate impurity, and are, generally speaking, harmless unless suffi- 
 cient numbers are present to affect the water by their death and 
 decay. (See also Chapter IV. — Surface water as a source of 
 supply.) 
 
 The second method of biological investigation was proposed 
 by Emmerich,* who says : " If the water under examination, or 
 the aqueous extract (of the residue), to the bulk of 40-80 c.c. be in- 
 jected subcutaneously into a full-grown rabbit, and fail to produce 
 a continued elevation of temperature of more than i" C, followed 
 by death, then the water contains no putrid substances dangerous • 
 to health, or contains them in so small an amount that they are 
 not worth considering." He further suggests, that the amount 
 of such dangerous substances may be estimated from the amount 
 of water which must be evaporated in order to obtain an extract 
 which will produce in animals the effects which he describes. 
 
 * Zeilschrift fur Biologie, xiv (1878), 563-603. 
 
46 WATER SUPPLY. 
 
 Something has been done in this same direction by Dr. George 
 M. Sternberg, U. S. A., * in connection with his investigation of 
 malarial fever, and also by Prof. H. Newell Martin, of the Johns 
 Hopkins University, in connection with Prof. Mallet's invest!-' 
 gation into the different methods for the analysis of drinking 
 waters. Whatever may develop in the future from this method 
 of research, it is certain that it has not yet become a method on 
 which we can rely, and that Emmerich's statement is altogether 
 too sweeping. 
 
 In this connection should also be mentioned the proposal by 
 Koch t to examine the water by means of culture experiments 
 in gelatine with subsequent microscopic examination ; but this 
 method, like the preceding, is at present simply developing. A 
 somewhat similar method of experimenting has, for a longer time, 
 been occasionally employed. It consists in carefully collecting 
 some of the water in flasks that have previously been sterilized 
 by heat and introducing a small quantity into each of several 
 test-tubes containing some freshly boiled Pasteur's solution, so- 
 called. This solution is particularly favorable to the growth of 
 " bacteria," if any such or their germs are in the water. The 
 tubes are plugged with cotton-wool, and an attempt is made to 
 estimate the amount of impurity in the water from the degree of 
 turbidity produced in a given time. None of these methods have 
 yet reached the stage of practical utility, and it must be left en- 
 tirely with the specialists to interpret the results of their own 
 observations. 
 
 Popular tests. — The writer has little sympathy with popular 
 tests. It is true that the observations on odor and taste and 
 color may be made by a person who is not a chemist ; there are 
 also certain qualitative tests that any intelligent person can learn 
 to make satisfactorily, and which would serve as indications to 
 the chemist. It is, in general, true of popular tests, that they 
 are apt to lead either to an unjustified sense of security or to an 
 unnecessary feeling of alarm. The following test for sewage 
 contamination, proposed by Heisch, and recommended by others, 
 has some value : 
 
 Put some of the water — say half a pint — into a clean, colorless, 
 
 * Supp. No. 14, Bulletin National Board of Health, July 23, 1881. 
 f Mittheilungen aus dem kaiserlichen Gesundheitsamte. I Band, Berlin, 1881, 
 page 36. See also a paper by Angus Smith, in The Sanitary Record, February, 1882. 
 
WATER ANALYSIS. 47 
 
 glass-stoppered bottle, add a few grains of white sugar, shake 
 until the sugar has dissolved, and leave the bottle freely exposed 
 to the light in a warm room for a week or ten days. If the 
 water becomes turbid, it is open to suspicion of sewage contam- 
 ination ; if it remains clear it is probably safe. 
 
 Collection of samples. — In connection with the chemical exam- 
 ination of water, the importance of taking due care in the collec- 
 tion of samples may be alluded to. The best vessel for collecting 
 water for analysis is a glass-stoppered bottle ; a clean demijohn 
 which has never been used for any other purpose and which is 
 stopped with a new and_ clean cork answers perfectly well and 
 is often more convenient. Tin cans or stoneware jugs are not 
 suitable. 
 
 Considerable care is necessary in order to get a fair sample of 
 the water. The demijohn should be rinsed several times thor- 
 oughly with the water to be collected and finally filled no^ quite 
 to the mouth. The cork should be washed with the same water 
 and the demijohn stoppered tightly. The stopper should be 
 tied over with a piece of cloth or " bandage gum," and the string 
 sealed with sealing "wax, that the water may not be tampered 
 with in transit. 
 
 If the water is taken from a pump or from a faucet, enough 
 water should be pumped or allowed to run to waste to thor- 
 oughly clear the pipes. In taking water fromi a pond or river 
 it will generally be most convenient to use a clean crockery 
 pitcher, which may be filled by plunging it beneath the surface 
 (so as to avoid any scum or floating material) and then emptied 
 into the demijohn; or a new and clean tin dipper may be 
 employed. If a glass bottle is used, it may be plunged directly 
 into the water and thus filled. In taking water from a river, the 
 middle of the stream should be chosen if only one sample is 
 taken. 
 
CHAPTER III. 
 
 RAIN WATER AS A SOURCE OF SUPPLY. 
 
 The collection of the rain directly as a source of public sup- 
 ply, in our latitude, would be undertaken only under very excep- 
 tional circumstances. In many localities, however, where there 
 is no sufficient public supply and where wells are out of the 
 question, the collection of rain water by the individual house- 
 holder becomes a necessity ; also in cases where the public supply 
 is hard and unfit for washing. A house covering an area of 40 by 
 20 feet, or 800 square feet, would receive upon its roof, with a 
 rainfSU of 42 inches, 2,800 cubic feet of water in the course of 
 the year. If three-quarters of this could be collected, it would 
 furnish a supply of something over 40 gallons per day, on the 
 average. 
 
 The rain which falls, even in the open country, is far from 
 being pure in the chemical sense, as it washes from the air both 
 gaseous and solid substances. The Rivers Pollution Commission 
 of Great Britain obtained the following average results from the 
 examinatidn of 73 samples of rkin water, all but two of which 
 were collected at the experimental farm of Messrs. Lawes and 
 Gilbert, Rothamste.d, England : 
 
 Total dissolved solids 3.95 parts in 100,000. 
 
 Organic carbon 0.099 
 
 Organic nitrogen 0.022 
 
 Ammonia '. o . 050 
 
 Nitrogen as nitrites and nitrates 0.007 
 
 Total comljined nitrogen 0.071 
 
 Clilorine 0.63 
 
 In manufacturing localities, the air, and consequently the 
 rain, may contain much impurity ; and, in any event, when the 
 rain is collected near habitations the impurity is considerable. 
 The excrement of birds, the dead bodies of insects, leaves from 
 the trees, and various sorts of dust and dirt lodge on the roofs 
 and are washed off, mainly by the first portions of the rain. 
 
RAIN WATER AS A SOURCE OF SUPPLY. 49 
 
 Devices, some of them automatic, have been invented, by which 
 the first portions of the rain are allowed to go to wa^ste, but they 
 have not come into general use. Besides the sources of impurity 
 to which rain water is naturally and unavoidably liable, there are 
 accidental sources of contamination : thus, instances have been 
 known where servants have emptied the house slops from the 
 upper stories on to the roof, thence to find their way into the 
 cisterns. It is to be hoped that such instances as this are ex- 
 tremely rare. 
 
 The proper storage of rain water is as important as its col- 
 lection. For the storage of water* in small quantity there is 
 nothing better, from a sanitary point of view, than slate tanks ; 
 iron tanks protected from rusting by a coating of coal-tar paint 
 are also unobjectionable. Tanks situated at the top of the 
 house are sometimes in direct communication with the drains 
 by means of the overflow pipes, and in some cases it has been 
 thought that the water has been rendered impure and injurious 
 to health by gases from the soil-pipes. Any mode of construc- 
 tion which admits the possibility of such communication is 
 faulty. Rain water is generally stored in underground cisterns 
 built of brick and cement, and acquires a slight hardness by 
 dissolving lime from the cement, especially when the cistern is 
 new. The overflows from such cisterns should be constructed 
 so as to preclude all possibility of contamination from the liquids 
 or gases of drains or sewers. In some localities wood is used 
 for the construction of the rain-water tanks ; thus in New 
 Orleans, according to Dr. Smart,* the cisterns are constructed 
 of cypress wood, and vary in capacity from 500 to 60,000 gallons. 
 "The usual capacity of the dwelling-house cistern is about two 
 thousand gallons. They are raised a few feet from the ground, 
 and their contents are protected by a lid or cover. Some are 
 placed under the shade of a balcony ; a few have a special roof 
 over them ; but the majority have only such protection from the 
 rays of the sun as is afforded by their position against the house 
 wall. Many, especially in the older parts of the city, are situated 
 in unventilated inclosures which are rank with the emanations 
 from unclean privies." 
 
 * Report on the Water Supply of New Orleans and Mobile. Bull. National 
 Board of Health, April 17, 1880. 
 4 
 
50 WATER SUPPLY. 
 
 On account of the sediment which accumulates in the cisterns 
 in which rain water is stored, it is desirable that such cisterns 
 should be thoroughly cleansed from time to time, and that the 
 water should be filtered before being used for drinking or for 
 culinary purposes. The matter of filtration will be discussed in 
 a subsequent chapter (page 151). Dr. Smart examined a number 
 of cisterns in New Orleans which had not been cleaned in many 
 years, and found that the rate of deposit was, on the average, 
 about an inch a year. After a few days' repose, the sediment 
 carried in by the rain has settled to the bottom and the water 
 has become clear, " but every succeeding rainfall not only in- 
 creases the quantity of this sediment, but, by its inflow, stirs up 
 that which has already accumulated, rendering the water impure 
 until sedimentation is again accomplished. As time passes the 
 sediment increases, and the water becomes unfit for use after 
 each rainfall. These conditions are aggravated in the dry season, 
 when the water is low in the cistern and the quantity of sedi- 
 ment is relatively much increased." 
 
 Underground cisterns, being out of sight and consequently 
 too often out of mind, are not only liable to be, neglected and 
 allowed to go uncleaned for a long time, but are also liable to 
 become leaky and thus to allow of the contamination of the 
 water by soakage from the soil. Dr. Smart, in his investigation 
 of the water supply of Memphis, examined a considerable num- 
 ber of the 4,000 cisterns said to be in use. He found — 
 
 Undoubtedly sound 163 
 
 Probably sound 82 
 
 Probably siping 94 
 
 Undoubtedly leaky igo 
 
 Total number examined 529 
 
 Examination of Cistern Water. 
 
 It is often very difficult to decide, by chemical examination, 
 whether a cistern water is to be regarded as fit to drink or not. 
 Gross pollution can be easily detected, but, as the presence of 
 more or less organic matter is a necessary consequence of the 
 mode of collecting the water, it is impossible to say just how 
 much is permissible, and where the line should be drawn. The 
 most valuable indications are afforded by the chlorine, which 
 
RAIN WATER AS A SOURCE OF SUPPLY. 
 
 SI 
 
 should be present only in trifling quantity, as a rule, not over 
 0.5 part in 100,000, except near the sea. The total solids should 
 not much exceed 4 or 5 parts in 100,000, but a larger amount is 
 sometimes due simply to the solvent action of the water on the 
 cement lining of the cistern. Dr. Smart considers cistern waters 
 containing from o.oio to 0.020 part of albuminoid ammonia in 
 100,000 as usable, and regards those containing over 0.020 part as 
 dangerous, but as much as o.oio ought to awaken suspicion and 
 give rise to inquiry. The following table contains the resuks of 
 the examination of a number of cistern waters in various localities. 
 
 TABLE V. — Examination of Cistern Waters. 
 [Results expressed in Parts in 100,000.] 
 
 Locality ■ 
 
 Oakham, Eng. . . . 
 
 Epsom, Eng 
 
 Goring, Eng 
 
 Podehole, Eng. . . 
 Sheffield Barracks. 
 Boston, Mass .... 
 Same, filtered*. . 
 
 Another 
 
 Same, filtered *. . 
 
 Another 
 
 Same, filtered *. . . 
 Wilmington, N. C 
 
 Another 
 
 Another 
 
 Cincinnati, O 
 
 Another 
 
 Another 
 
 Another 
 
 Another 
 
 (/I 
 
 < 
 
 i 
 
 s 
 
 s 
 <! 
 
 
 
 
 § 
 
 6 
 
 11.70 
 
 0.005 
 
 
 3331 
 
 8.10 
 
 0.009 
 
 
 3.167 
 
 9.42 
 
 0.032 
 
 
 3.215 
 
 5.28 
 
 0. 
 
 
 3.142 
 
 12.00 
 
 0.130 
 
 
 3.154 
 
 S-28 
 
 0.013 
 
 0.008 
 
 
 6.56 
 
 0.012 
 
 0.007 
 
 
 
 3.24 
 
 0.005 
 
 O.OII 
 
 
 
 4.80 
 
 0.024 
 
 0.016 
 
 
 
 3.48 
 
 0.02 [ 
 
 0.007 
 
 
 
 5.20 
 
 0.007 
 
 0.007 
 
 
 
 5.05 
 
 0.002 
 
 0.015 
 
 
 
 6.90 
 
 o.or6 
 
 0,008 
 
 
 
 3.60 
 
 0.005 
 
 0.008 
 
 
 
 2.68 
 
 0.004 
 
 0.123 
 
 
 
 4.72 
 
 0275 
 
 0.055 
 
 
 
 4.48 
 
 0.027 
 
 0.II8 
 
 
 
 7.96 
 
 0.004 
 
 o,ot6 
 
 
 
 4.10 
 
 0.020 
 
 0.360 
 
 
 
 < O 
 o « 
 
 o 
 
 0.090 
 0.021 
 0.061 
 0.029 
 
 0.053 
 
 0.90 
 0.80 
 
 o.go 
 1.60 
 0.32 
 0.36 
 o.io 
 0.12 
 0.69 
 0.70 
 0.70 
 0.52 
 0.52 
 
 0-55 
 2.76 
 1.97 
 trace 
 trace 
 
 Authority. 
 
 Riv. Poll. Com. I 
 
 W. R. Nichols t 
 
 C. W. Dabney§ 
 
 C. R. Stuntzll 
 
 The table might be extended indefinitely, but the results have 
 little real significance except in conhection with a minute knowl- 
 edge of the history of the various samples. 
 
 While rain water, on account of its softness, is peculiarly 
 
 * These cisterns were provided with a brick filtering-wall. 
 
 f Sixth Report, p. 29. 
 
 i Filtration of Potable Water, p. 83. 
 
 § Report North Carolina Agricultural Experiment Station, 1881, p. 158. 
 
 II Report of W.-iter Department, Cincinnati, 1880, p 80. 
 
52 WATER SUPPLY. 
 
 adapted to use in washing and in cooking, it is also wholesome 
 as a beverage if collected so as to be reasonably pure. There 
 has, however, long been an opinion that snow water is unwhole- 
 some. Dr. Charles Brewer, U. S. A., says that " mountaineers 
 (in the West), to whose long observation and experience in the 
 wilds some attention is due, attribute the origin of the so-called 
 mountain fever to the melting of snows and the drinking of snow 
 water." Dr. Smart * quotes facts to show that this suspicion is 
 well grounded, and -says that " snow water, therefore, pure as it 
 seems, must not be accepted as innocent until its freedom from 
 organic ammonia in deleterious or suspicious quantity Jias been 
 proved." This applies especially to snow melting and flowing in 
 the mountain streams. As collected with the rain and stored in 
 cisterns it is wholesome, as far as we know. (Compare p. loo). 
 
 NATURAL AND ARTIFICIAL ICE. 
 
 As the rain is water which has been more or less completely 
 purified by a natural process of distillation and condensation, so 
 ice is water which has been more or less completely purified by 
 another natural process — that of crystallization. It is, therefore, 
 not inappropriate to consider ice in this connection. 
 
 The great extent to which ice is used in many parts of the 
 United States makes its purity a matter of very considerable 
 importance. The surface water supplies, even of Northern cities, 
 become heated in summer so that, to cool the water, at least 
 half its own weight of ice is needed, and very much more than 
 this is often used in practice. 
 
 When ice forms in or on a body of water, the bulk of which 
 remains unfrozen, it is generally supposed that foreign sub- 
 stances are excluded and that the ice itself is essentially pure. 
 It is true that foreign, especially saline, substances are excluded 
 to a large extent, but the ice always retains more or less, and, 
 if the water contains organic impurities, the ice will contain them 
 also. In fact, a pond or river which is not fit for a water supply 
 on account of present evident contamination should not be used 
 as a source of ice s'upply. It is also, unfortunately, the case that 
 ice is often cut in winter on shallow .ponds which for a consider- 
 
 * Buck's Hygiene, ii, p. 133. 
 
DANGERS OF IMPURE ICE. 53 
 
 able portion of the year have no existence or exist merely as 
 stagnant pools. That such ice should be wholesome, ought not 
 to be expected, and there are well authenticated cases on record 
 where sickness has arisen from the use of impure ice. Such a case 
 of well-marked character occurred in 1875 at Rye Beach, N. H.* 
 The ice had been cut on a brackish stagnant pond into which a 
 small brook brought a quantity of saw-dust from several saw- 
 mills. Here, the trouble was ascribed to the presence of decom- 
 posing organic matter. That, in some cases, the germs (if they be 
 germs) of specific diseases might retain their vitality even if 
 frozen into the ice, we can but regard as possible from what we 
 know of the endurance of the spores of the lower algse ; at any 
 rate, ice should not be cut on contaminated waters. 
 
 It is unfortunately true that when public attention is called 
 to a possible, but hitherto little noticed cause of disease, exag- 
 gerated statements are sure to be disseminated through the 
 public prints and otherwise — statements which have a basis of 
 truth, but which are so presented as, oftentimes, to awaken un- 
 necessary anxiety and alarm. The statement — " The old idea 
 that water purifies itself by freezing is now pretty generally 
 abandoned " — is true, if it means that water does not thus purify 
 itself completely ; it is untrue as far as it naturally leads to the 
 inference that water, in freezing, does not purify itself at all. 
 Further, the author believes that the following statements (in 
 which the italics are his), even if hereafter proved true, are cer- 
 tainly premature, and have not been proved by experiments 
 hitherto published. " The even more dangerous organic im- 
 purities resulting from human and other animal waste are retained 
 in ice unchanged as regards both quality and quantity, the latter 
 indeed being likely to be increased." " The germs of infectious 
 disease . . . are retained in ice unaffected, and /rojn their 
 comparative lightness -are so concentrated therein as to number that 
 they exist in even greater quantity than in the same amount of 
 water, under similar circumstances, at other seasons of the year." f 
 
 Large quantities of artificial ice are now made in the United 
 States, and in a number of Southern cities the artificial product 
 
 * See a paper by A. H. Nichols, M. D., in the Seventh Report of the Mass. State 
 Board of Health, 1876, p. 465. 
 
 \ The Dangers of Impure Ice in The Sanitarian for May, 1882. See also a paper on 
 Impure Ice, in the Fifth Annual Report of the Connecticut State B6ard of Health, 1883. 
 
54 WATER SUPPLY. 
 
 has driven the natural ice from the market. With some of the 
 machines, distilled water alone is frozen ; with others, ordinary 
 well or other water. Ice machines, on a small scale, are used 
 also in hotels for freezing carafes, etc. Particular care should be 
 used with reference to the water employed in making artificial 
 ice, because the water is frozen solid, and whatever is dissolved or 
 suspended in the water must remain in the ice. In one machine, 
 however (Beath's patent), the ice is formed by causing water to 
 flow over pipes through which the freezing agent flows. Only a 
 part of the water used is actually frozen as it flows over the 
 pipes or the continually increasing thickness of ice, and the bulk 
 of the impurities, dissolved or suspended, flows away. 
 
 Chemical Examination of Ice. 
 
 The standard of purity for ice should be placed very high. 
 Ice should contain very little dissolved matter, next to no chlo- 
 rine, and the "albuminoid ammonia" should not exceed 0.005 
 part in 100,000. Great care must be exercised in preparing the 
 sample for analysis, because the ice in melting attracts organic 
 impurities from the air. 
 
 In fact, one method which has been proposed for examining 
 the air for organic matter, consists in employing a glass funnel 
 drawn to a point and filled with fragments of ice.* The moisture 
 in the air condenses as dew upon the outside of the funnel, 
 trickles down into the vessel below, and the 
 water thus collected is examined for organic 
 matter in various ways. For this reason, also, 
 the fact that the drip from refrigerators often 
 becomes offensive when allowed to stand in a 
 warm room, does not show necessarily that 
 the ice used was impure. 
 
 In melting ice for analysis, a fair specimen 
 cake should be selected and broken into frag- 
 ments in a clean place. The fragments may 
 then be placed in a wide-mouthed bottle cov- 
 ered with a plate of glass, and, when enough 
 Fig. 5. Qf j-jie j^e has melted to have washed itself, 
 
 * Smee : Social Science Transactions, 1875, p. 486. The figure is from Fox's 
 Sanitary Examination of Water, etc. 
 
CHEMICAL EXAMINATION OF ICE. 
 
 55 
 
 this portion of water should be poured away, and the remainder, 
 after melting, be subjected to analysis. Or the fragments may 
 be placed in a large funnel covered with a plate of glass until 5 
 or 10 per cent has melted, and the remainder be then transferred 
 to the wide-mouthed bottle without touching the ice with the 
 hands. Since even the best of ice is liable to have bits of organic 
 substances frozen into it here and there, the " albuminoid am- 
 monia " should be determined also in the water which has been 
 filtered through paper. It must also be remembered that the 
 ice which is sold for family use often is partly " snow ice," and 
 that the snow, in falling, always brings down ammonia from the 
 atmosphere. Table VI contains the results of a number of 
 chemical examinations of ice from various sources. 
 
 TABLE VI. — Examination of (Melted) Ice. 
 [Results expressed in Parts in ioo,cxxj.] 
 
 Description. 
 
 Rye Beach, N. H. Supposed cause of sickness 
 
 Same, filtered through paper 
 
 Boston Ice Company, 1875 
 
 Fresh Pond, near Boston, 1878 
 
 Spy Pond, ^' " " 
 
 Little Spy Pond, near Boston, 1878 
 
 HornPond, " " 1876* 
 
 Hammond's Pond, '' " 1877 
 
 Jamaica Pond Ice Co. , 1877 
 
 Another sample of same 
 
 Pittsfield, Mass., 1876 
 
 Same, filtered through paper 
 
 s 
 
 0,0208 
 0.0213 
 0.0045 
 0.0060 
 0.0064 
 0.0060 
 0.0026 
 0,0066 
 0.0180 
 0.0260 
 0.0072 
 0.0077 
 
 
 
 
 z ^ 
 
 u 
 
 
 ss 
 
 
 Authority. 
 
 " s 
 
 
 
 
 f< 
 
 X 
 
 u 
 
 
 . 0704 
 
 
 W. R, Nichols. 
 
 0.0165 
 
 3-2 
 trace 
 
 u 
 
 0.0075 
 
 
 S. P. Sharpies. 
 
 0.0064 
 
 
 " 
 
 O.OIIO 
 
 
 " 
 
 . 0440 
 
 0.4 
 
 E. S. Wood. 
 
 o.oigo 
 
 
 
 0.0160 
 
 0.3 
 
 " 
 
 0160 
 
 0.3 
 
 " 
 
 0.0061 
 
 
 W. R. Nichols. 
 
 0.0013 
 
 0.02 
 
 
 * The large amount of impurity in this case was probably due to some local and 
 accidental cause : this ice was thought to cause sickness. 
 
CHAPTER IV. 
 
 SURFACE WATERS AS SOURCES OF SUPPLY. 
 
 The character of the water which flows in the streams and 
 is stored in the lakes and ponds of any region depends largely 
 upon the geological character of the country. Where impervious 
 rocky strata predominate, the water flows off from the surface 
 readily, and the streams carry water free from any very consider- 
 able amount of dissolved substances : where the water soaks 
 quickly into the ground and the streams and lakes are fed by 
 springs, or, at least, by water which has passed for some distance 
 underground, the amount of dissolved matter becomes more con- 
 siderable. As a rule, the surface waters contain less dissolved 
 matter and are softer than the well waters (ground water) of the 
 same region ; on the other hand, they generally contain more 
 organic matter of animal and vegetable origin, they are often 
 colored, they are liable to be turbid or to become so at time of 
 flood, and are, in several respects, more subject to variation than 
 the ground water. 
 
 While much that might be said with reference to surface 
 waters will apply both to running streams and to ponded waters, 
 it will be convenient to divide the subject, discussing first the 
 matters of turbidity and pollution, which are peculiarly felt in 
 the case of rivers, and subsequently discussing the difficulties 
 arising from variable temperature and from animal and vegetable 
 growths, which are felt peculiarly in the case of ponded waters. 
 
 Turbidity of Streams. 
 
 It is hardly necessary to dwell upon the fact that rivers are 
 frequently objectionable as sources of supply, on account of the 
 large amount of suspended matter, mainly clay, which many 
 of them carry invariably and others at time of flood. Some- 
 thing will be said of this matter in connection with sedi- 
 mentation in Chapter VIII: at present, it will suffice to call 
 
THE POLLUTION OF STREAMS. 
 
 57 
 
 attention to Table VII (made up from statements in Geikie's 
 Text-book of Geology) which shows the amount of suspended 
 matter in various rivers, and also shows that the amount varies 
 very much in the same stream under different circumstances. 
 
 TABLE VII. — Amoutw of Suspended Matter in Various Rivers. 
 
 River. 
 
 Rhone .... 
 Rhone . 
 
 Rhine 
 
 Maas 
 
 Danube — 
 
 Vistula 
 
 Po 
 
 Durance..., 
 
 Ganges 
 
 Irrawaddy. 
 
 Mississippi 
 
 Date. 
 
 1844. 
 
 Dec. 1 
 
 1862-1; 
 
 849 
 
 871 
 
 Locality and Condition. 
 
 Lycns , 
 
 Aries , low water 
 
 '' flood 
 
 " maximum 
 
 " mean 
 
 Strasburg, July and Aug 
 
 Bonn 
 
 Bonn 
 
 Bonn, after dry weather 
 
 Uerdingen, after sudden floods, 
 
 In Holland 
 
 Maximum 
 
 Minimum 
 
 Mean 
 
 Mean 
 
 Maximum 
 
 In flood 
 
 Mean 
 
 In flood 
 
 Yearly mean , 
 
 In flood , 
 
 Low water 
 
 Mean , 
 
 Suspended 
 
 Parts of 
 
 
 Matter 
 
 Water for 
 
 
 Expressed in 
 
 I part of 
 
 
 Parts in ioo,ocx). 
 
 Sediment. 
 
 Authority. 
 
 by 
 
 by 
 
 V^, 
 
 bv 
 
 
 weight. 
 
 volume. 
 
 w'g't. 
 
 vol. 
 
 
 5.88 
 
 
 17,000 
 
 
 
 14.29 
 
 
 7,000 
 
 
 
 434.78 
 
 
 
 230 
 
 
 
 2,222.22 
 
 
 45 
 
 
 
 50,00 
 
 
 2,000 
 
 
 
 2,00 
 
 
 50,000 
 
 Daubrde. 
 
 
 6.25 • ■■ 
 
 i6,'.oo Horner. 
 
 20.50 
 
 
 4,878 
 
 .... Bischof. 
 
 1-73 
 
 
 57,800 
 
 
 78.00 
 
 
 1,282 
 
 Stiefensand. 
 
 
 T,000.00 
 
 
 100 Hartsoeker. 
 
 47.61 
 
 
 2,100 
 
 
 Chandellon. 
 
 1 .40 
 
 
 71,420 
 
 
 '* 
 
 10.00 
 
 
 10,000 
 
 
 '^ 
 
 32,68 
 
 17.20 
 
 3,060 
 
 5,814 Hartley. 
 48 Spittell. 
 
 
 2,083.00 
 
 
 
 333-00 
 
 
 300 Lombardini. 
 
 2,083,33 
 
 
 48 
 
 .... iPayen. 
 
 100, OD 
 
 
 1,000 
 
 .... 1 " 
 
 233-9S 
 
 116.82 
 
 4?8 
 
 856 Everest. 
 
 jg6.o8 
 
 97.94 
 
 510 
 
 X,02I 
 
 
 58,82 
 
 
 1,700 
 
 
 Login. 
 
 17-47 
 
 
 5>72S 
 
 
 66.66 
 
 34.48 
 
 1,500 
 
 2,900 
 
 Humphreys 
 and Abbot. 
 
 TAe Pollution of Streams. 
 
 Owing to facilities for transportation, to available water- 
 power, and also to the opportunities furnished for the discharge 
 of waste material, running streams naturally attract to their 
 banks manufactories and towns, and, in turn, become polluted 
 by their refuse. 
 
 In a thickly settled manufacturing country like England, 
 where, moreover, the streams are comparatively small, the pollu- 
 tion may become very serious. In Great Britain the matter has 
 been the subject of thorough investigation by two Royal Com- 
 missions, appointed respectively in 1865 and 1868, "to inquire 
 into the best means of preventing the pollution of rivers." 
 
 The statement of the Commissioners with reference to the 
 Aire and Calder, although it has been frequently quoted, has not 
 lost in emphasis : 
 
58 WATER SUPPLY. 
 
 " The rivers Aire and Calder and their tributaries are abused 
 by passing into them hundreds of thousands of tons per annum 
 of ashes, slag and cinders from steam-boilers, furnaces, iron- 
 works and domestic fires ; by their being made the receptacle, 
 to a vast extent, of broken pottery and worn-out utensils of 
 metal, refuse brick from brick-yards and old buildings, earth, 
 stone and clay from quarries and excavations, road-scrapings, 
 street-sweepings, etc. ; by spent dye-woods and other solids used 
 in the treatment of worsteds and woollens ; by hundreds of car- 
 casses of animals, as dogs, cats, pigs, etc., which are allowed to 
 float on the surface of the streams or putrefy on their banks ; 
 and by the flowing in, to the amount of very many millions of 
 gallons per day, of water poisoned, corrupted, and clogged by 
 refuse from mines, chemical works, dyeing, scouring and fulling 
 worsted and woollen stuffs, skin-cleaning and tanning, slaughter- 
 house garbage, and the sewage of towns and houses." 
 
 " Bradford is an ancient town situated on a ' beck ' about four 
 miles south of the river Aire, into which the water of this beck 
 falls. It is the center of the worsted district." The Commis- 
 sioners allude to the increase of population and to the increased 
 pollution from dye-works, from soap-suds, and from refuse of 
 various kinds produced in manufactures. " The whole of the 
 sewage of Bradford, and of the populous district above the town, 
 flows into the beck, producing an indescribable state of pollu- 
 tion. It has become a Yorkshire proverb of comparison for any 
 foul stream, to say of it that it is as polluted as Bradford Beck. 
 At the time of our inquiry Bradford Beck was the source of 
 supply of the Bradford Canal, the fluid of which became so cor- 
 rupt in summer that large volumes of inflammable gases were 
 given off, and, although it has usually been considered an impos- 
 sible feat ' to set the River Thames on fire,' it was found practi- 
 cable to set the Bradford Canal on fire, as this at times formed 
 part of the amusement of boys in the neighborhood. They 
 struck a match placed on the end of a stick, reached over, and 
 set the canal on fire, the flames rising six feet high and running 
 along the surface of the watei- for many yards like a will-o'-the 
 wisp ; canal boats have been so enveloped in flame as to frighten 
 persons on board." 
 
 The river Irwell, near its source, " is of excellent quality for 
 all domestic purposes." It flows, however, through the midst of 
 
THE POLLUTION OF STREAMS. 59 
 
 a manufacturing district, and finally passes through Manchester 
 before it empties into the Mersey. At Manchester the slug- 
 gishly flowing stream is black as ink, and it is there joined by the 
 Irk and Medlock, streams not less polluted than itself. 
 
 Of the Clyde, the following are bits of evidence : 
 
 " At one time the Clyde was comparatively pure and limpid — 
 salmon fishing within the precincts of the harbor being very pro- 
 ductive. Now, through Glasgow downwards, but diminishing 
 below the mouth of the Cart, it is very foul and turbid ; in short ■ 
 a gigantic open sewer, noxious gases being continually evolved, 
 which, during summer, are so overpowering as to force the bulk 
 of the passenger trafific from the river to the rail." 
 
 " The harbor is more like a gigantic cesspool than a harbor in 
 the proper acceptation of the term." 
 
 " In summer time there is a perfect commotion with air and 
 gas bubbles over the whole surface of the water, and it is so bad 
 that we cannot use it for the boilers of the little steam ferry- 
 boats that ply across the river." 
 
 While the rivers of Great Britain are probably polluted gen- 
 erally to a greater extent than those of most other countries, the 
 trouble is by no means peculiar. In France the condition of the 
 Seine below Paris has led to the appointment of several depart- 
 mental and municipal commissions, and to the proposal of exten- 
 sive and costly plans for disposing of the sewage of the city. 
 Other rivers of France, as, for instance, the Vesle, at Rheims, 
 have become the receptacles of town and manufacturing refuse 
 so as to call imperatively for restrictive action. In Germany the 
 increasing pollution of the sluggish Spree by the sewage of 
 Berlin was one of the moving causes which has led to an entire 
 change of the system of sewerage, and to the attempt at purifi- 
 cation and utilization of the entire sewage of the city on sewage 
 farms. 
 
 In this country, many of our streams carry such a volume of 
 water that the refuse of the largest cities is soon lost ; but some 
 of the smaller rivers, for a portion of their course at any rate, 
 are rendered unfit for domestic use. The Blackstone River, in 
 Massachusetts, receives the sewage of Worcester and causes 
 complaint from the towns and manufactories on its banks. The 
 Schuylkill and Passaic rivers are no longer fit for water supply 
 where water-works now exist. The Chicago River was an ex- 
 
6o WATER SUPPLY. 
 
 ceedingly foul stream — if stream it could be called — until it was. 
 diverted into the Illinois River ; and other examples might be 
 cited. The contamination of the water of the Great Lakes, in 
 the neighborhood of cities like Cleveland, Chicago and Milwau- 
 kee, is a very serious matter. There is a limit to the distance to 
 which tunnels can be carried into the open lake, and the problem 
 of disposing of the sewage of the cities, otherwise than by dis- 
 charging it into the lake, is one which will soon compel solution. 
 
 It is not our purpose to enter into details as to the nature 
 and amount of the polluting materials which are discharged by 
 various manufacturing industries, nor to discuss how far the indi- 
 vidual substances are injurious to plants or to fish, or how far 
 they render the water unfit for drinking and for other domestic 
 purposes.* We may admit that certain substances are injurious, 
 even in small amount; but, while this is true, it is also true 
 that much manufacturing refuse is of such a character as to 
 be, except in excessive quantities, of no appreciable influence on 
 the human system. Thus, the addition to a water of most of 
 the ordinary salts of lime, soda, potash, etc., would not produce 
 any deleterious effect, although the addition of lime compounds 
 would increase the hardness and render the water less desirable 
 for washing. Again, in the case of many waste liquors of offen- 
 sive appearance, the actual amount of matter which is really 
 injurious or of suspicious character is comparatively small. Thus, 
 in the case of some of the organic dyestuffs, the weak, spent 
 dye-liquors, although they communicate a very foul appearance 
 to the water for some distance, yet contain a comparatively small 
 amount of solid matter, and, if discharged into a stream of con- 
 siderable size, are soon disseminated through it, and diluted to a 
 very great extent. 
 
 Very different, however, in character and importance, from 
 much of the refuse of manufacturing establishments,, is, as we 
 have seen, the sewage proper — that is, the excremental matters 
 from factories and towns — and the refuse from particular opera- 
 tions, such as tanning, slaughtering, rendering, wool-pulling, etc. 
 
 * These matters are very fully discussed in the reports of the Rivers Pollution 
 Commissions of Great Britain. A brief statement with reference to the chemicals and 
 other materials used in various manufacturing operations and with reference to the 
 liquid refuse discharged from them, is given in the Report of the Mass. State Board 
 of Health, 1876 : Special Report on the Pollution of Rivers by J. P. Kirk wood, C.E. 
 
THE POLLUTION OF STREAMS. 
 
 6l 
 
 With our present information, too much stress cannot be laid 
 upon the importance of preventing the discharge of such refuse, 
 and of sewage in its more restricted sense, into any stream or 
 pond used, or Ukely to be used, as a source of water supply. 
 
 The importance of this matter is underrated for two reasons : 
 first, because of a belief that an impure and polluted water rap- 
 idly purifies itself by natural means ; and, second, because of the 
 feeling that a water to be prejudicial to health must be polluted 
 to such an extent that the animal matter may be recognized by 
 chemical tests. 
 
 That a polluted water in its flow does become purer, no one 
 can doubt who has followed the course of a polluted stream ; 
 chemical analysis proves the same thing. There is, however, 
 much difference of opinion as to the method by which the puri- 
 fication takes place, and also as to the extent to which we may 
 suppose that the disease-producing something is eliminated. 
 
 TABLE VIII. — The Seine above and below Paris.* 
 
 Kilometers. 
 
 31 
 
 78 
 
 93 
 
 109 
 
 150 
 
 242 
 
 Locality. 
 
 Corbeil (above Paris) 
 
 Pont de la Tournelle, Paris 
 
 Auteuil (below the city but above 
 
 the outlets of the main sewers) . . . 
 
 Pont d'Asnieres (above main sewer) 
 
 Epinay (below all sewers) 
 
 Pont de Poissy 
 
 Pont de Meulan 
 
 Mantes 
 
 Vernon 
 
 Rouen 
 
 Organic 
 
 Total 
 
 Dissolved 
 
 Nitrogen. 
 
 Combined 
 
 Oxygen. 
 
 
 Nitrogen. 
 
 Cubic 
 
 Grams per 
 
 Grams per 
 
 centimeters 
 
 cubic meter. 
 
 cubic meter. 
 
 per liter. 
 
 .. 
 
 
 9-32 
 
 •• 
 
 
 8.05 
 
 
 
 5-99 
 
 0.85 
 
 1.9 
 
 5-34 
 
 1.26 
 
 3.0 
 
 1.05 
 
 0.45 
 
 2.2 
 
 6 12 
 
 0.40 
 
 
 8.17 
 
 •• 
 
 1-4 
 
 8.96 
 10.40 
 10.42 
 
 Table VIII contains the results of partial examinations of 
 the Seine above and below Paris. At Epinay, below all the 
 sewers, the river is at its worst as regards the amount of nitrogen 
 in the form of ammoniacal salts and organic compounds, and the 
 dissolved oxygen is reduced to a minimum. After flowing some 
 75 or 100 kilometers, the river regains its purity as far as appear- 
 ance and chemical tests can indicate. 
 
 * Assainlssement de la Seine, etc., deuxi^me partie, II Annexes, p. 8 ; also Rap- 
 port de MM. Schloesing et A. Durand-Claye. Congres international d'hygiene, 
 Paris, 1878, p. 314. 
 
62 
 
 WATER SUPPLY. 
 
 TABLE IX. — Self-Purification of Streams. 
 [Hesults expressed in Parts in 100,000.] 
 
 Miles. 
 
 5 
 25 
 
 I3i 
 
 o 
 
 Locality. 
 
 Blacks tone River, 1875. 
 
 North Pond, above pollution 
 
 City Reservoir, gate-house 
 
 Mill Brook, below sewers 
 
 Blackstone River, at sash factory 
 
 Blackstone River, at Blackstone 
 
 Merrimack River, 1873. 
 
 Mean of 11 examinations above IjOwell. . . 
 Mean of 12 examinations above Lawrence. 
 Mean of 1 1 examinations below Lawrence 
 
 Merrimack River, 1879.* 
 
 Mean of 2 examinations above Lowell. . . . 
 Mean of 4 examinations above Lawrence . 
 
 Total 
 Solids. 
 
 Ammonia. 
 
 " Albuminoid 
 Ammonia." 
 
 Chlo- 
 rine. 
 
 4.20 
 
 0.0107 
 
 0.0213 
 
 0.18 
 
 3.76 
 
 0.0072 
 
 0.0235 
 
 0.12 
 
 23 -44 
 
 0.9600 
 
 O.II09 
 
 3.80 
 
 8.04 
 
 0.0992 
 
 0.0307 
 
 0.92 
 
 4.80 
 
 0.0099 
 
 0.0139 
 
 0.36 
 
 4.10 
 
 0.0047 
 
 O.OI14 
 
 0.14 
 
 4.10 
 
 0.0044 
 
 O.OIIO 
 
 0.20 
 
 4-43 
 
 0.0031 
 
 0.0127 
 
 0.18 
 
 5-50 
 
 0.0021 
 
 0.013,2 
 
 0.40 
 
 7-50 
 
 0.0018 
 
 0.0I3I 
 
 0.44 
 
 Table IX contains results which are less striking but which 
 point in the same direction. The Blackstone Valley is the seat 
 of considerable manufacturing industries, there being on the 
 stream and its tributaries 44 woolen mills, 27 cotton mills, 12 iron 
 works, I tannery and i slaughter house. Some of the mills 
 cause local pollution, but the chief source of contamination is 
 the sewage of the City of Worcester — some 2,000,000 gallons in 
 24 hours — which flows into Mill Brook and thence into the 
 river. 
 
 " The water of Mill Brook, after it has received the sewage of 
 Worcester, is shown to be very impure in this table, and on the 
 Blackstone River, at the sash factory, about five miles lower 
 down, it still gives unmistakable signs of the influence of this 
 pollution ; but at Blackstone, twenty-five miles below Mill Brook, 
 the dilution produced by numerous small streams delivering into 
 the main river between these points has all but obliterated the 
 evidence of impurity, so far as analysis can expose it, the only 
 marked, difference here in the table between the water at Black- 
 stone and the head water of the river, being in the amount of 
 chlorine, the increase, however, of this evidence of impurity not 
 being so great as to condemn the water (by this test) for 
 domestic or any other use. It is to be noted, however, that the 
 river at this time was not at its very low dry-weather stage, 
 
 * E. S. Wood, M.D. 
 
THE POLLUTION OF STREAMS. 63 
 
 which usually occurs in October or November, when it occurs at 
 all. In extreme low water, the river would give greater tokens 
 of impurity." * 
 
 Of the Merrimack River it may be said that, in 1873, when 
 some of the examinations were made, Lowell had a population 
 of about 41,000 and Lawrence of about 30,000; further, at 
 Lowell there are some 75 mill buildings, in which about 16,000 
 operatives are employed. About 10,000 horse power is derived 
 from the river, and, in addition, steam power is used to the extent 
 of 6,000 horse power. The Merrimack Manufacturing Company 
 alone consumes, among other things, 7,500 gallons of oil per 
 annum, 225,000 pounds of starch, 1,100 barrels of flour, 2,500,000 
 pounds of madder, 50,000 of copperas, 170,000 of alum, 200,000 
 of sumac, 1,120,000 of sulphuric acid, 300,000 of bark, 350,000 of 
 soda-ash, and 40,000 of soap.f 
 
 At Lawrence there are some 25 mills (buildings), employing 
 9,000 operatives. The manufacturing industry is less at Law- 
 rence than at Lowell, but it is still very considerable. The 
 Pacific Mills, which is the largest corporation, use some 800,000 
 pounds of starch, 540 barrels of flour, 8,300 gallons of oil, etc.:]: 
 
 The questions naturally arise, to what causes are we to ascribe 
 the disappearance of the large amount of polluting material in the 
 Seine, and in the Blackstone, and why, in the case of the Merri- 
 mack River, are we not able to trace a greater effect as produced 
 by the large manufacturing cities of Lowell and Lawrence. 
 
 In studying the self-purifying power of streams, let us first 
 take an instance of a substance whose course we can trace more 
 easily than that of animal refuse, with which we are, of course, 
 more concerned. The following account from the First Annual 
 Report of the Mass. State Board of Health, Lunacy and Charity 
 (1880), will furnish the illustration : 
 
 "On the night of June 2, 1879, a fire occurred in a chemical 
 works situated on a brook whose waters eventually find theirway 
 into Mystic Pond, from which a portion of the city of Boston, 
 Mass., obtains its supply of water. As a result of the destruc- 
 
 * Seventh Annual Report of Mass. State Board of Health, 1876, p. 84. 
 
 f These figures are taken from the " Statistics of Lowell Manufactures, January, 
 l873>" published by Stone & Huse, Lowell. 
 
 X " Statistics of Lawrence Manufactures, January, 1872." Published by Geo. S. 
 Merrill & Co., Lawrence. 
 
64 
 
 WATER SUPPLY. 
 
 tion by fire of the sulphuric acid chambers, a considerable quan- 
 tity of sulphuric acid, estimated at fifty tons of oil of vitriol, 
 together with salt cake and other chemicals, was washed directly 
 into the brook, or flowed upon the adjoining meadow-land, from 
 which it would slowly find its way to the stream. Large num- 
 bers of fish, driven before the acid water, or actually killed by it, 
 passed into the mill-ponds below and through the wheels of the 
 mills. Anxiety was felt lest the acid should reach Mystic Pond 
 itself ; and, five days after the fire, specimens of the wa^ef were 
 collected for analysis. As far as Mystic Pond itself was con- 
 cerned, the fears proved groundless; but in the brook and in some 
 of the upper ponds there was an abnormal amount of dissolved 
 matter and especially of sulphates. The most interesting point, 
 however, was with reference to the acidity of the water. As a 
 rule, our surface waters in Massachusetts are naturally slightly 
 alkaline, and, when the water is evaporated to dryness, the residue 
 effervesces, at least slightly, when treated with acid.* It was 
 found that even five days after the fire the water of the brook 
 itself and of the nearest ponds was distinctly acid. The amount 
 of acid was estimated by means of a dilute solutioft of baryta, 
 using rosolic acid as an indicator. 
 
 " The acidity was found to be as follows, the results being 
 expressed by stating how many parts by weight of sulphuric acid 
 
 
 Date. 
 
 Locality. 
 
 Acidity. 
 
 No. 
 
 2 = 
 
 Q 
 
 as 
 
 
 
 
 
 '"S 
 
 
 
 • 
 
 ¥^ 
 
 l^ 
 
 
 1879. 
 
 
 
 
 I 
 
 une 7, A.M. 
 
 From brook just below works 
 
 1.74 
 
 57.500 
 
 IV 
 
 . une 7, A.M. 
 
 Lower end of Richardson's Pond, about ilf miles below 
 
 
 
 
 
 works ■ 
 
 0.74 
 
 135,000 
 
 V 
 
 June 7, A.M. 
 
 Frye & Thompson's Pond, about 3 miles below works. . . . 
 
 0.37 
 
 270,000 
 
 11 
 
 June 7, P.M. 
 
 From brook midway between Chemical Works and Rich- 
 
 
 
 
 
 ardson's Pond, about }i mile below works 
 
 From canal at Montvale, about jX miles below works 
 
 Upper end of Richardson's Pond, >i mile below 11 
 
 0.18 
 
 S5S.S0O 
 
 VI 
 
 June 7. p.M 
 
 0.37 
 
 270,000 
 
 III 
 
 June 8, p.M 
 
 0.15 
 
 666,600 
 
 (H2SO4), or its equivalent, were present in 100,000 parts by weight 
 of the water ; and also by stating, in round numbers, with how 
 
 * Whether the alkaUnity is to be regarded as due in part to the presence of alka- 
 line carbonates, or as solely due to the presence of dissolved carbonate of calcium, is 
 uncertain, as there are no analyses which are sufficiently particular to determine. 
 
SELF-PURIFICATION OF STREAMS. 
 
 65 
 
 many parts of water by weight one part of sulphuric acid was 
 diluted. 
 
 " From the point numbered VI, that is from the canal at 
 Montvale, samples were taken at intervals until the water re- 
 turned to its alkaline reaction. The results of the examination 
 were as follows : 
 
 Date. 
 
 1879. 
 June 6 
 
 June 7 
 
 June 9 
 
 June 10 
 
 June 13 
 
 June IS 
 
 June 19 
 
 June 30 
 
 July 8 
 
 
 
 
 rs-o' 
 
 
 
 
 
 
 
 
 <.s^ 
 
 
 Li^L) 
 
 > 
 
 u-^O 
 
 
 So 
 
 3 
 
 .as"i 
 
 Locality. 
 
 .38 
 
 8-. 
 
 00' 
 
 |8« 
 
 ■fg-s 
 
 =^ rt C 
 
 
 -■s 
 
 ^ 
 
 w " 5 
 
 
 
 ■3— 
 
 
 glg 
 
 
 H 
 
 <! 
 
 ^H 
 
 Canal at Montvale 
 
 j.ii.6 
 
 
 
 Avenue, Montvale 
 
 
 
 
 13.9 
 
 0.37 
 
 
 
 12.8 
 
 0.26 
 
 
 
 II. 6 
 
 0.22 
 
 
 
 9.4 
 
 0,16 
 
 
 
 10, 1 
 
 J Slightly 
 I alkaline. 
 
 }■■■■ 
 
 
 9-4 
 
 o.i6 
 
 1.2 
 
 
 7.7 
 
 Sliffhtly 
 alkaline. 
 
 
 
 
 8.4 
 
 Slightly 
 alkaline. 
 
 
 
 Remarks. 
 
 The residue of evaporation did not 
 f" effervesce with acid. 
 
 Residue did not effervesce. 
 Residue did effervesce with acid. 
 
 " While the water was acid, the residue of evaporation did 
 not effervesce with acid, showing that a part, at any rate, of the 
 sulphuric acid was neutralized by the carbonates in the water. 
 It may also be noted, that some fragments of marble were put 
 into some of the water 'No. IV, which had an acidity of 0.74; 
 after standing for fifteen hours the acidity had decreased to 
 o.ii, and after standing for two and a half days the reaction was 
 neutral or faintly alkaline." 
 
 Mystic Pond is about eight miles below the works : samples 
 were taken from the upper end of the pond for several days, but 
 no acid reaction was at any time perceptible. 
 
 Thus, in the disappearance of the sulphuric acid, we have 
 two actions concerned : first, the chemical action by which it 
 was converted probably into sulphate of lime, and then the 
 action of dilution, by which all traces of it were apparently lost. 
 With many inorganic substances we may predict what will hap- 
 pen when they come into a stream, and the same thing is true 
 
66 WATER SUPPLY. 
 
 of certain particular organic substances whose changes have 
 been studied under various conditions ; but of the heterogeneous 
 mixture which we speak of collectively as " sewage," it would be 
 difificult to declare a priori what changes would take place and 
 what products would be formed. There are, however, three 
 principal ways in which a natural water frees itself from organic 
 pollution: (i) by oxidation and other chemical changes; (2) by 
 deposition ; (3) by dilution. 
 
 Oxidation. — The disappearance of organic refuse is, no doubt, 
 to a certain extent, due to chemical changes, by which the or- 
 ganic matter is oxidized and converted into simpler compounds; 
 under favorable circumstances the destruction may be complete, 
 the nitrogen appearing in ammoniacal compounds and nitrates, 
 or escaping in the free state, and the carbon appearing as car- 
 bonic acid. That chemical action takes place is made evident, 
 enough, when the pollution is considerable, by the escape of sul- 
 phuretted hydrogen, marsh gas, and other gaseous products of de- 
 composition, as in the case of the Bradford Beck — alluded to on 
 page 58. Of the Seine, it was said in 1874 that, for the greater 
 part of the year, especially in warm weather, bubbles of gas — ' 
 sometimes a meter or a meter and a half in diameter — were 
 continually rising to the surface of the water, and the passage of 
 a boat caused "a great ebullition, and Jeft a mass of foam persist- 
 ing for some minutes. The chemical change is also marked by 
 the partial or total disappearance of dissolved oxygen, as in- 
 stanced in Table VIII. 
 
 Attempts have been made to reach, by laboratory experi- 
 ments, some idea of the amount of oxidation which may take 
 place in a running stream polluted by sewage. The Rivers 
 Pollution Commission mixed urine with water in the propor- 
 tion of one gallon of urine to 3,077 gallons of water. The 
 mixture was agitated from time to time, and samples taken for 
 analysis. The results (expressed in parts in 100,000) were as 
 follows : 
 
 Immediately after mixture, Feb. 
 
 
 
 Organic 
 Carbon. 
 
 Organic 
 Nitrogen. 
 
 17. 
 
 1874 
 
 0.282 
 
 0.243 
 
 18, 
 
 
 0.298 
 
 L1.251 
 
 19, 
 
 
 0.244 
 
 0.255 
 
 24, 
 
 
 225 
 
 0.253 
 
 25. 
 
 
 0.214 
 
 0.259 
 
 28, 
 
 
 0.214 
 
 0.276 
 
SELF-PURIFICATION OF STREAMS. tj 
 
 In other experiments, a stream of impure water was allowed 
 to flow from one vessel to another, freely exposed to the air. 
 As a general result, the commissioners concluded that the puri- 
 fication due to actual oxidation had been much overrated, and 
 that there was no river in the United Kingdom long enough, if 
 once seriously polluted, to purify itself in this way. 
 
 Certain chemical changes, other than those already alluded 
 to, will be mentioned in the next paragraph : we may here 
 allude to the fact that in the destruction of organic matter a 
 certain part is, no doubt, played by the fish and other animal 
 inhabitants of the water, and many of the changes which seem 
 to be purely chemical appear to be brought about wholly or in 
 part by some of the lower algie or by those minute organisms 
 which are frequently spoken of together as "bacteria." Thus 
 we know there are certain algae which reduce the sulphates to 
 sulphides, and the formation of nitrates in the soil and in water 
 is ascribed also to micro-organisms. 
 
 Deposition. — Much waste material thrown into rivers is made 
 up wholly or in part of substances insoluble in water. A portion, 
 and a very considerable portion, even in a running stream, is 
 deposited upon the bottom or stranded upon the banks. This 
 deposition can often be very plainly observed in the immediate 
 neighborhood of the points of discharge. It is not, however, 
 suspended matters only that are removed by deposition. In the 
 first place, many substances which seem to be perfectly dissolved 
 are -dragged out of solution and carried down by almost any 
 finely divided substance. This is especially true of organic 
 coloring matters and of the nitrogenous matter which gives rise 
 to the " albuminoid ammonia " revealed by analysis ; it is true, 
 but to a very limited degree, of some mineral salts. In the 
 second place, chemical changes take place in the stream, espe- 
 cially when refuse liquids from different sources meet and mix, 
 which result in the formation of new and insoluble substances, 
 and these when formed settle out, dragging other substances 
 with them as just indicated. The improvement in the con- 
 dition of polluted streams which appeals to the eye is largely 
 due to deposition. 
 
 The deposits, once formed, continue to undergo chemical 
 change ; they shift their position with changing currents, and in 
 time of flood may be washed up and, mingling with earthy 
 
68 WATER SUPPLY. 
 
 material held in suspension, be swept on to the sea or deposited 
 in some new position, either lower down on the stream or on 
 the surface of overflowed territory. Often, however, the character 
 of the bed of the stream becomes permanently altered unless 
 means are taken by dredging ot otherwise to remove the ac- 
 cumulations as they increase. 
 
 Dilution. — Probably the most important reason of the appar- 
 ent disappearance of sewage and other waste material, is the fact 
 that the amount of solid matter is so small compared with the 
 volume of water into which it is thrown, that it is disseminated 
 through the mass, and thus lost to observation, and, in many 
 cases, to chemical tests. 
 
 A river in its flow loses water by evaporation, by being di- 
 verted for manufacturing and other purposes, and in some places, 
 through crevices in a rocky bed or by a slower process of per- 
 colation if the bed be gravelly. On the other hand, the stream 
 receives water from tributaries more or less pure than itself, 
 and its volume is further increased by the entrance of ground 
 water or by actual springs rising i(i its bed. On this account we 
 cannot calculate the extent to which dilution will take place with 
 any great accuracy, but that the apparent disappearance of much 
 of the foreign matter which finds its way into the stream is 
 really owing to dilution is evident when we undertake to trace 
 the course of some substance not liable to undergo any changes 
 which will result in its actual disappearance. The chlorine, 
 which exists in the form of chloride of sodium (common salt) 
 and other chlorides, furnishes us with what we need. 
 
 All natural waters contain a small proportion of chlprine, very 
 small in inland waters, slightly increased in waters near the sea. 
 Chlorine is a. universal accompaniment of sewage, generally in 
 the form of chloride of sodium (common salt), and occurs also in 
 most manufacturing refuse. All the chlorine used in the process 
 of bleaching is eventually washed away, and that contained in 
 the various compounds of this element which are used in dye- 
 houses and print-works finds its way in the end into the drains 
 of the establishments. On this account, although harmless in 
 the combinations in which it finally exists, its presence indicates 
 the presence, now or formerly, of refuse material. Of course, in 
 regions containing salt-springs and salt-deposits, these statements 
 would require modification. It is to be remarked further, that 
 
SELF-PURIFICATION OF STREAMS. 69 
 
 while there do exist compounds of chlorine which are insoluble 
 in water, and other compounds which are gaseous, and while it is 
 true that chlorides, are absorbed to a limited extent by earth 
 filters and by growing plants, for all practical purposes we 
 may say that the chlorides once dissolved in the water are not 
 removed either as insoluble or gaseous compounds. 
 
 If, now, we take the case of the Blackstone; River as instanced 
 in Table IX, page 62, we shall see that the 3.80 parts of chlorine 
 in Mill Brook have become 0.92 part five miles below, after the 
 brook has been merged in the river, and twenty miles further 
 still the river water contains only 0.36 part. Believing, as we do, 
 that this decrease of the chlorine is due almost entirely to dilu- 
 tion, we must believe also that the decrease in the amount of the 
 organic matter is largely due to the same cause. Owing to the 
 fact that the organic matter is much more liable to conversion 
 into insoluble and volatile compounds, no one would deny that 
 an appreciable amount of organic matter is chemically changed 
 and actually destroyed, but emphasis should be laid upon what 
 we believe to be a fact — namely, that the apparent self-purifica- 
 tion of running streams is largely due to dilution, a?id the fact that 
 a river seems to have purified itself at a certain distance below a 
 point where it was certainly polluted, is no guaranty that the 
 water is fit for domestic use. 
 
 Referring again. to Table IX, it would appear that no effect 
 is produced upon the waters of the Merrimack by the cities of 
 Lowell and Lawrence, which throw into the stream the greater part 
 of their sewage and the waste from all the mills. Is the organic 
 matter of all this refuse destroyed by oxidation or other chem- 
 ical change ? Certainly not ! Let us take simply the results of 
 examination above and below Lawrence. Between these two 
 points the river receives the refuse from nearly all the manufac- 
 turing establishments, a large proportion of the excreta of the 
 factory operatives, and a portion of the sewage of Lawrence. 
 Moreover, the lower station is so short a distance below the city, 
 that no chemist, probably, would believe that any considerable 
 destruction of organic matter could take place in the rapid flow 
 for so short a distance, and if, upon chemical grounds, the evi- 
 dence was not sufificient, the floating soap-suds, with still unbroken 
 bubbles, and other materials borne down upon the current show 
 the same thirig. Now, in spite of this addition, which must be 
 
70 WATER SUPPLY. 
 
 considerable, there is only a slight increase in the total dissolved 
 solids, and in the albuminoid matters, while the chlorine is prac- 
 tically the same or slightly decreased. We know, positively, that 
 chlorine compounds, in large quantity, are thrown into the 
 river at Lawrence, and yet, just below the city, the proportion is 
 not increased. Although absolutely large, the amount is too 
 small compared with the great volume of water to produce an 
 appreciable increase. From these considerations with reference 
 to the chlorine, it is evident that, in the case of the soluble 
 organic matter it is not necessary to suppose any destruction or 
 decomposition ; the apparent decrease or lack of increase may be 
 explained, as in the case of chlorine, by the fact of dilution, and 
 where the distance between the two points of examination is so 
 short as in the instance now under discussion (above and below 
 Lawrence), this is no doubt the main cause concerned. 
 
 It may be well, in this connection, to say that it is often diffi- 
 cult to prove satisfactorily the actual pollution of a stream until 
 by the accidental or unusual discharge of some peculiarly offen- 
 sive or characteristic substance the matter is placed beyond 
 doubt. Carbolic acid has sometimes served this purpose ; thus, 
 a few years ago, the taste and odor of carbolic acid were so strong 
 in the water supplied to Newark, Jersey City and Hoboken, 
 that the water could not be used for domestic purposes. These 
 cities take their water from the Passaic River at a point where 
 the stream is always polluted, but at this time particular atten- 
 tion was called to the polluted condition of the water supply 
 by the washing of a quantity of carbolized paper at a mill on 
 one of the tributaries of the stream.* The following circum- 
 stance is related with reference to the water supply of Cincin- 
 nati, Ohio, f 
 
 " The eddy of the Deer Creek Canal caused much pollution, 
 which was brought to the attention of the consumers in a 
 marked manner in 1867 by the burning of a large distillery on 
 the canal. The whiskey found its way into the canal, thence 
 into the river, and thence to the consumers." 
 
 Much depends, of course, upon the size of the stream into 
 which the refuse is thrown. Thus, while into the Merrimack at 
 
 * Leeds : Journ. Amer. Cliem. Soc, , iii. 
 f Engineering News^ 1881, p. 162. 
 
PREVENTION OF POLLUTION. 7 1 
 
 Lowell, even during the summer, it would be necessary to throw 
 more than one hundred tons of solid matter daily in order to 
 increase the amount in the water by one grain to the gallon,* 
 another and smaller stream might be hopelessly fouled by a 
 single factory. 
 
 Ullik has made some interesting calculations with reference to 
 the Elbe, in Bohemia, and shows that the entire product of the 
 well-known sulphuric acid works at Aussig, if allowed to flow 
 into the river, would increase the amount of sulphates by only 
 one twenty-fourth, and that the mineral matter in the sewage 
 from all of the 5,000,000 inhabitants of Bohemia would increase 
 the mineral matter in the stream by only one twentieth. f 
 
 A question which we should be glad to have answered is this : 
 To what extent must a polluted liquid be diluted in order to be 
 safely used for domestic purposes ? The answer, however, none 
 can give. We do know this : it has been shown by actual exper- 
 iment that the spores of some of the lower orders of vegetable 
 organisms are very difificult to deprive of vitality ; they may be 
 frozen or heated to the boiling temperature, or they may be kept 
 in a dry condition for years, and then, if placed in a favorable 
 medium, become active and produce their kind. Admitting the 
 presence of disease germs in a liquid, the liquid may be diluted 
 until the charrce of taking even a single germ into the system is 
 so small that it may be disregarded ; and yet, if the prevailing 
 theory be true, a single germ, if taken, might produce disastrous 
 results. It is easy to push the demands for purity to an absurd 
 extent ; all reasonable precautions should be taken to insure 
 purity, but there is a point beyond which it is foolish to attempt 
 to go. In the present state of our knowledge we should, how- 
 ever, err on the side of safety, and the mere fact that chemical 
 analysis fails to detect impurity should not be accepted as a 
 guaranty that a water is fit to drink. 
 
 Prevention of Pollution. 
 
 The pollution of streams can be prevented only by legal 
 enactments. Cities and towns claim the right to discharge their 
 
 * Fifth Annual Report of the Mass. State Board of Health. 
 t Abh. d. math, wissenschf. Classe d. k. bohm. Ges., x. 
 
72 WATER SUPPLY. 
 
 sewage into a water-course on which tliey may be situated, and 
 unless a nuisance is thereby created within their own boundaries, 
 they are not Hkely, of their own motion, to do anything toward 
 a different disposal of the sewage, as all other methods involve 
 additional expense. Riparian owners, especially when they have 
 located manufactories for the sake of the conveniences which 
 the stream affords, will not, as a rule, adopt any method for dis- 
 posing of their sewage or manufacturing refuse which involves 
 expense, unless compelled to do so. Even where waste material 
 may be utilized, so that the expense is met by a corresponding 
 return, the vis inertice is so great that legal pressure is generally 
 necessary to secure the adoption of plans which, in the end, may 
 prove as advantageous to the manufacturer as to his previously 
 injured neighbor. 
 
 The following is a very brief summary of the legal restric- 
 tions which exist in several European States : * 
 
 " In Prussia, though there are no special regulations for the 
 inspection of factories in order to prevent the pollution of run- 
 ning waters, there are very simple means of interference in every 
 separate case of pollution by manufacturing refuse. Thus, by 
 the statute of July I, 1861, most of the processes which are 
 likely to pollute water require a special police license ; and by a 
 statute passed as far back as 1843, no water applifed to dyeing, 
 tanning, fulling, or other similar purposes, is to be suffered to 
 enter a river, if thereby the means of procuring clean water be 
 endangered to the neighborhood ; and by a regulation dating 
 from, October 28, 1846, the owners of such works as impregnate 
 the water used in any manufactory, with materials hurtful to 
 meadow land, must, in accordance with the judgment and direc- 
 tion of the police authorities, precipitate their materials in sub- 
 sidence ponds, or otherwise, under penalty of fine. 
 
 " In Belgium there are various local regulations, having the 
 force of law, which impose a penalty upon those who pollute 
 rivers, either by throwing in solid materials, which may impede 
 the course of the stream, or by allowing liquid matter, which may 
 foul or corrupt the water, to flow into it. Manufactories which 
 produce such refuse are bound to construct reservoirs sufficiently 
 large to contain a day's supply of this refuse, in order that suf- 
 
 * Rivers Pollution Commission. First Report 1870, p. 42. 
 
PREVENTION OF POLLUTION. 73 
 
 ficient settlement may take place, and the supernatant liquid only 
 is allowed to be run off." 
 
 In France, several royal ordinances and decrees of the Coun- 
 cil (1669, 1672, 1773, 1777, 1783), forbid the pollution of streams. 
 All of these ordinances and decrees, which still have the force of 
 law, prohibit, under penalty, casting into the Seine or into other 
 water-courses, " aucunes ordures, immondices, gravois, pailles et 
 fumiers." The laws of December 22, 1789, and August 16-24, 
 1790, grant to departmental and municipal authorities power to 
 preserve the purity of streams, and to interpose if the waters 
 become a source of ill-health. A ministerial decision, dated July 
 24, 1875, reaffirms, in principle, the decrees of 1773 and 1777.* 
 
 That the laws, even of well-policed countries like Germany 
 and France, do not succeed in wholly accomplishing the object 
 desired, is evident from the present condition of certain streams, 
 notably of the Seine and Spree. In England, there were formerly 
 Jaws similar to those existing in France ; but the enormous de- 
 velopment of the manufacturing industry in that country, and 
 the national sensitiveness as to the liberty of the subject, served, 
 in course of time, to render them all dead letters. 
 
 The Public Health Act, of i848,tgave local authorities power 
 to build sewers and discharge them into streams wherever they 
 saw fit. Then began, on a large scale, the pollution of the rivers 
 of the country which has since become so great an evil. By the 
 Nuisances Removal Act, 1855, provision was first made for en- 
 joining individuals, towns and corporations against the pollution 
 of streams. To provide for the full carrying out of the require- 
 ments of the act. Section IX contains a clause that the local 
 authorities shall appoint or employ a sanitary inspector or in- 
 spectors. In 1858, local authorities were rendered liable to 
 injunction for polluting streams. 
 
 In the Sewage Utilization Act of 1865, we find this clause; 
 "Nothing contained in this act or any other acts referred to 
 therein {i. e., all the sanitary acts previously passed), shall author- 
 ize any sewer authority to make a sewer so as to drain direct into 
 any stream or water-course." By the same acts the powers of 
 sewer authorities were extended as to the pollution of streams. 
 
 * Schloesing et Durand-Claye. Rapport. 
 
 ■f This sketch of English legislation is condensed from the Seventh Annual 
 Report of the Mass. State Board of Health. 
 
74 WATER SUPPLY. 
 
 Seventeen years had sufficed to reverse entirely the laws on the 
 subject. In 1848, towns were urged to empty sewage freely into 
 the most convenient water-courses. 
 
 The first Rivers Pollution Commission was appointed in 1865 ; 
 the second was appointed in 1868, and their last report (the sixth), 
 was printed in 1874. Largely as a result of the careful and com- 
 prehensive investigations of these commissions, and especially of 
 the second one, there was passed, in 1876, a Rivers Pollution 
 Prevention Act,* which was certainly a great step in advance, 
 although the act is not satisfactory to sanitarians in all respects. 
 The provisions of the act are, briefly, as follows : f 
 
 I. Prohibition as to casting solid matter (ashes, dead animals, 
 etc.) into water-courses. 
 
 II. Prohibition as to casting sewage proper into water- 
 courses. In case, however, of sewage discharged by channels in 
 use or process of construction at the date of passage of the act, 
 it will be sufficient to show that the best practicable and avail- 
 able means are used to render harmless the sewage so discharged, 
 and the Local Government Board may allow to sanitary author- 
 ities time, in order to adopt such means. 
 
 III. (i) Prohibition as to casting poisonous, noxious, or 
 polluting refuse from manufactories into water-courses, with the 
 same provision as above with regard to channels already in use 
 or in process of construction. Proceedings against manufactories 
 can be taken only by consent of the Local Government Board, 
 who must be satisfied that means for rendering the manufactur- 
 ing refuse harmless are reasonably practicable and available, 
 under all the circumstances of the case, and that no material 
 injury will be inflicted upon the interests of the manufacturers. 
 
 (2) Restrictions as to solid matter from mines. 
 
 The following definition was formulated by the Rivers PoL 
 lution Commission (1868) of liquids which should be deemed 
 polluting and inadmissible into any stream, but they have not 
 been established by legal enactment. 
 
 (a) Any liquid which has not been subjected to perfect rest 
 in subsidence ponds of sufficient size for a period of at least six 
 hours, or which having been so subjected to subsidence^ contains 
 
 * The text of this act may be found in the Eighth Annual Report of the Mass. 
 State Board of Health, p. 73. 
 
 f Quoted from Ninth Annual Report of the Mass. State Board of Health. 
 
PREVENTION OF POLLUTION. 75 
 
 in suspension more than one part by weight of dry organic 
 matter in 100,000 parts by weight of the hquid, or which, not 
 having been so subjected to subsidence, contains in suspension 
 more than three parts by weight of dry mineral matter, or one 
 part by weight of dry organic matter in 100,000 parts by weight 
 of the Hquid. 
 
 {b) Any hquid containing, in solution, more than two parts 
 by weight of organic carbon or 0.3 part by weight of organic 
 nitrogen in 100,000 parts by weight. 
 
 {c) Any liquid which shall exhibit by daylight a distinct color 
 when a stratum of it one inch deep is placed in a white porcelain 
 or earthenware vessel. 
 
 {d) Any liquid which contains in solution, in 100,000 parts 
 by weight, more than two parts by weight of any metal except 
 calcium, magnesium, potassium, and sodium. 
 
 {e) Any liquid which in 100,000 parts by weight contains, 
 whether in solution or suspension, in chemical combination or 
 otherwise, more than 0.05 part by weight of metallic arsenic. 
 
 (/) Any liquid which, after acidification with sulphuric acid, 
 contains, in 100,000 parts by weight, more than one part by 
 weight of free chlorine. 
 
 {g) Any liquid which contains, in 100,000 parts by weight, 
 more than one part by weight of sulphur, in the condition either 
 of sulphuretted hydrogen or of a soluble sulphuret. 
 
 {h) Any liquid possessing an acidity greater than that which 
 is produced by adding two parts by weight of real muriatic acid 
 to 1,000 parts by weight of distilled water. 
 
 (t) Any liquid possessing an alkalinity greater than that which 
 is produced by adding one part by weight of dry caustic soda to 
 1 ,000 parts by weight of distilled water. 
 
 {k) Any liquid exhibiting a film of petroleum or hydrocarbon 
 oil upon its surface, or containing in suspension, in 100,000 
 parts, more than 0.05 part of such oil. 
 
 To these standards was attached the proviso, that " no 
 effluent water shall be deemed polluting if it be not more 
 contaminated with any of the above-named polluting ingredients 
 than the stream or river into which it is discharged." 
 
 In this country comparatively little has been done in the way 
 of legislation from a sanitary point of view, as the polluted 
 streams are still few in number, and the necessity for legislation 
 
76 WATER SUPPLY. 
 
 has not become pressing. In some States there are general laws 
 against the obstruction and pollution of water-courses, or special 
 provision for securing, to a certain extent, the purity of streams 
 or other waters actually used as sources of supply. 
 
 In the District of Columbia there is a law of the United 
 States (1859), which provides penalty for committing any act by 
 reason of which the supply of water to the cities of Washington 
 and Georgetown becomes impure, filthy, or unfit for use. In 
 loiva, by an act of 1864, it is punishable by fine or imprison- 
 ment to throw any dead animal into any river, well, spring, 
 cistern, reservoir, stream, or pond. In Michigan, an act of 1865 
 prohibits the putting of offal, etc., into waters where fish are 
 taken. In Nebraska, by act of 1873, there are penalties for 
 putting carcasses or other filthy substances into well, spring, 
 brook, or any running water of which use is made for domestic 
 purposes ; the corrupting of any water-course is declared to be 
 a nuisance. 
 
 In Tennessee, by act of 1866-7, rendering water unwhole. 
 some is declared a nuisance. In Vermont, by act of 1852, a 
 penalty was enacted against any one putting any dead animal 
 or animal substance into rivers, ponds, springs, etc. In Wis- 
 consin there are laws providing against the erection of slaughter- 
 houses on the banks of any river, stream or creek, or throwing 
 any carcass or offal therefrom in or upon the bank of any such 
 river, etc. In Texas, by act of i860, polluting or obstructing 
 any water-course, lake, pond, marsh, or common sewer, or con- 
 tinuing such obstruction or pollution, so as to render the same 
 unwholesome or offensive to the county, city, town, or neighbor- 
 hood thereabouts, or doing any other act or thing that would be 
 deemed and held a nuisance at common law, is made a misde- 
 meanor.* 
 
 The Texas law, just alluded to, indicates very well what is 
 generally the character and effect of legislation on this subject. 
 If any person or corporation pollutes a stream in such a way 
 that the result would be held a nuisance at common law, or if 
 the pollution is so great that it ca,n be absolutely proved that 
 water, which is taken for domestic use, has become " impure, 
 
 * These facts are gathered from "A Digest of American Sanitary Law, by Henry 
 G. Pickering, Esq.," in Dr. Bowditch's Public Hygiene in America. Boston, 1877. 
 
PREVENTION OF POLLUTION. jy 
 
 filthy, or unfit for use," it is, under such circumstances, possi- 
 ble in some cases to obtain an injunction against the offending 
 parties. But to absolutely prove that a water is impure, filthy 
 or unfit to use, is difficult unless one can present in court the 
 body of some person who has died by the use of the water ; 
 and, according to certain decisions which have been made, it 
 would seem that nothing short of this will suffice. In several 
 States a local or a State Board of Health has some powers in 
 these matters, but they are mostly nominal and rendered of no 
 avail by rights of appeal. A great deal of attention has been 
 given to the pollution of streams in the State of Massachusetts, 
 and the two following sections from chap. 183 of the Acts of 1878 
 would seem sufficiently explicit : 
 
 "Section i. No person or persons, or corporation, public 
 or private, shall discharge directly, or cause to be discharged 
 directly, human excrement into any pond in this Commonwealth 
 used as a source of water supply by any city or towk therein, or 
 upon whose banks any filter-basin so used is situated, or into 
 any river or stream so used, or upon whose banks such filter- 
 basin is situated, within twenty miles above the point where such 
 supply is taken, or into feeders of such ponds, river or stream 
 within such twenty miles. 
 
 " Section 2. No person or persons, or corporation, public or 
 private, shall discharge, or cause to be discharged, into any 
 pond in this Commonwealth used as a source of water supply by 
 any city or town therein, or upon whose banks any filter-basin 
 so used is situated, or into any river or stream so used, or upon 
 whose banks such filter-basin is situated, within twenty miles 
 above the point where such supply is taken, or into any feeders 
 of such pond, river, or stream within such twenty miles, any sew- 
 age, drainage, refuse, or polluting matter of such quality and 
 amount, as either by itself, or in connection with other such matter, 
 shall corrupt or impair the quality of the water for domestic use, 
 or render it deleterious to health." * 
 
 It would seem that these provisions would give all necessary 
 security, but two practical difficulties occur. Section 3 reads as 
 follows ; 
 
 "The prohibitions contained in the two previous sections 
 shall not be construed to destroy or impair rights already ac- 
 
 * The italics are the author's. 
 
78 WATER SUPPLY. 
 
 quired by legislative grants, or to destroy or impair prescriptive 
 rights of drainage or discharge, to the extent to which they 
 lawfully exist at the date of the passage of this act : And nothing 
 in this act contained shall be construed to authorize the pol- 
 lution of any waters in this Commonwealth in any manner now 
 contrary to law. 
 
 " This act shall not be applicable to the Merrimack or Con- 
 necticut rivers, nor to so much of the Concord River as lies 
 within the limits of the city of Lowell." 
 
 The deciding whether a prescriptive right exists is not always 
 easy, but the chief difificulty lies in the fact that while the State 
 Board of Health is given the power to issue orders "to cease 
 and desist," the parties have a right of appeal to jury. This 
 jeopardizes the whole matter at once, and, as the pollution 
 often takes place in one county to the detriment or supposed 
 detriment of water-users in another county, the jury are apt to 
 be influenced by local interests. 
 
 While it must be confessed that the present state of legisla- 
 tion is unsatisfactory, it must be admitted that it is practically 
 impossible absolutely to prevent the pollution of streams, and it 
 will probably always be necessary that certain streams should 
 serve as carriers of refuse. At the same time, the amount of 
 pollution should be kept within bounds, so that the water, 
 although it may not be fit or safe for domestic use, shall not be 
 an actual nuisance ; while some streams are thus devoted to viler 
 uses, those which are reserved for purposes of water supply 
 should be guarded with all possible care. It is evident that 
 nothing is more unphilosophical than that one town should be 
 allowed to discharge its sewage into a water-course which is the 
 most available source of water supply for a town lower down on 
 the stream. Each river basin should be under the control of 
 some central authority by which conflicting interests should be 
 harmonized. An accurate survey should be made of the whole 
 area, and no town should be allowed to introduce a water sup- 
 ply without due consideration being given to the future of the 
 supply, and to the question of disposing of the sewage of the 
 town supplied. Moreover, while sanitary considerations are of 
 the highest importance, manufacturing interests must also be 
 considered, and no undue burden laid upon legitimate indus- 
 tries. 
 
CHAPTER V. 
 
 SURFACE WATERS AS SOURCES OF SUPPLY. {Continued^) 
 Animal and Vegetable Life. 
 
 No surface water is free from various forms of animal and 
 vegetable life ; it is seldom, however, that any trouble arises from 
 this cause when the water is taken directly from running streams. 
 On the other hand, the water in lakes and ponds and in storage 
 reservoirs is extremely liable to be disagreeably affected by the 
 growth and decay of animal and more especially of vegetable 
 organisms. 
 
 Animals. — The presence of fish in a source of supply is an 
 advantage rather than otherwise. As far as known, any trouble 
 arising from their presence is quite temporary and accidental. 
 The sudden discharge into a stream of material injurious in itself, 
 or which, by using up the oxygen of the water, makes it impos- 
 sible for the fish to breathe, may cause their destruction in large 
 numbers. Occasional epidemics, in some cases due to fungous 
 growths, may affect large numbers of fish at the same time. In 
 such cases, the sources of supply must be watched and the dead 
 fishes removed as thoroughly and as rapidly as possible. 
 
 Among the smaller forms of animal 
 life we have the so-called water-fleas, of 
 which the Daphnia pulex (Fig. 6) and 
 the Cyclops quadricornis (Fig. 7) are fa- 
 miliar examples. These or 
 other similar animalcules 
 swarm in many surface waters 
 at certain seasons of the year, 
 and occur more or less abun- 
 dantly in all pond and river 
 waters. They no doubt, to a 
 certain extent, tend to purify 
 the water by removing objec- 
 tionable substances, and are in turn devoured by other animals. 
 
 Fig. 6. — Daph- 
 nia pulex. 
 
 Fig. 
 
 — Cyclops quadri- 
 cornis. 
 
8o WATER SUPPLY. 
 
 They are easily removed by the simplest sort of filtration, but 
 there is no probability that they are unwholesome if taken with 
 the water. These minute crustaceans secrete an oily substance 
 under their shells (carapaces) which some have held to be the 
 cause of certain bad tastes which have affected water supplies, 
 but it is extremely doubtful if this is a correct explanation of 
 the trouble. It may be said, however, that a fishy odor has been 
 frequently noticed on the Lake of Geneva, very perceptible to 
 the passengers on the passing steamboats, occasionally over the 
 entire lake. This odor is ascribed by Dr. F. A. Forel * to an 
 unusual mortality^ — from some unknown cause — among the en- 
 tomostraca which swarm in the lake, and which usually by day 
 descend to a depth of 5, 10, or 20 meters, rising to the surface at 
 night. It is possible that some of the temporary bad tastes in 
 surface waters may be due to a similar cause. 
 
 In 1881, a portion of the water supply of Boston, Mass., was 
 in a very bad condition. The water contained an unusual 
 amount of organic matter and possessed a very disagreeable 
 odor and taste. This bad condition of the water was found by 
 Professor Remsen,f to be mainly due to the presence in one of 
 the reservoirs of a large quantity of a Spongilla, or fresh-water 
 sponge, in a more or less decayed condition. Several species of 
 the fresh-water sponge occur in ponds and streams, and they 
 are found even in the masonry conduits to a limited distance 
 from the source of supply. The following cut will give a fair 
 idea of the general appearance of the Spongilla fluviatilis, al- 
 though the details are somewhat unsatisfactory. The sponge 
 belongs to the animal kingdom, and the animal substance is dis- 
 tributed over a network of spicules. No. 2 in the figure at- 
 tempts to show some of the " winter buds," or bodies by which 
 the animal is propagated, held in a mass of spicules ; No. 3 is a 
 portion of the same enlarged. The sponge is harsh to the touch, 
 and with a good lens something of its structure may be made 
 out ; for confirmation, however, it is well to burn off a little on 
 a fragment of mica, moisten with water (or better, with acid, 
 muriatic acid or dilute aqua fortis) and examine with a good lens 
 
 * Private communication. 
 
 \ Report of the Joint Standing Committee on the Impurity of the Water Supply. 
 Boston City Document, No. 143, 1881. 
 
ANIMAL AND VEGETABLE LIFE. 
 
 8l 
 
 or a low-power microscope (say from 50 to 100 diameters) for the 
 spicules. The larger spicules, which are observed by examining 
 with a low power the sponge of the Boston water supply, appear 
 as in Fig. 9 (drawn with 
 a power of about 1 50 di- 
 ameters). Other spic- 
 ules are covered with 
 short spines, arid the so- 
 called winter-buds, by 
 which the sponge is 
 propagated, are furnish- 
 ed with minute spicules 
 which are, in certain 
 species, very character- 
 istic : thus in S. fluviat- 
 ilis they are birotulate, 
 as shown in Fig. 10, a. 
 Sorne of the smaller 
 spicules of the sponge 
 which occurs in the Bos- 
 ton water are shown in 
 Fig. 10, b and c. Fig. 
 10 was drawn with a 
 power of about 260 di- 
 ameters. 
 
 Although the spon- 
 ges have caused trouble 
 in Boston and perhaps 
 elsewhere, the fact that 
 they exist in a water 
 supply need not neces- 
 sarily awaken appre- 
 hension. They prob- 
 ably exist in nearly 
 all surface waters, and the spicules are among the common 
 objects found by a microscopic examination of the sediment 
 in natural waters.* According to Dr. R. D. Thompson, the 
 
 *See for exampb, the plates in Hassall's Microscopical Examination of the 
 Water supplied to the Inhabitants of London and the Suburban Districts. London, 
 1831. Also, Neuville, Des Eaux da Paris. Paris, 1880. 
 
 6 
 
 Fig. 8. — Spongilla Jiuviatilis^ 
 
82 
 
 WATER SUPPLY. 
 
 dried Spongilla fliiviatilis contains 26 per cent of organic matter. 
 
 This organic matter is higlily nitrogenous, althiough no particular 
 
 analyses seem to have 
 been made of the fresh wa- 
 ter sponges. The tissue- 
 substance of salt water 
 sponges — fibroin (Mulder) 
 or spongin (Stadeler) — 
 has the following compo- 
 sition : 
 
 
 Posselt. 
 
 Crookcwit 
 
 Carbon . . . 
 
 48.75 
 
 47.16 
 
 Hydrogen 
 
 6-35 
 
 6.31 
 
 Nitrogen . 
 
 16.40 
 
 16.15 
 
 Oxygen . . 
 
 28. 50 
 
 26. yo 
 
 Iodine . . . 
 
 
 I.oS 
 
 Sulphur . . 
 
 
 0.50 
 
 Phosphorus 
 
 
 I. go 
 
 
 100.00 
 
 100.00 
 
 Fig g. — Sponge spicules. 
 
 Among the other forms of animal life which may be here 
 mentioned, is the Hydra, which is common enough in ponds, 
 adhering to aquatic plants. The body is narrow and elongated ; 
 it is generally attached 
 by the base to some 
 plant or other solid ob- 
 ject while the other ex- 
 tremity of the body is 
 furnished with long slen- 
 der arms or tentacles, 
 which move about in 
 search of the animalcules 
 on which it feeds. It 
 may be worth while to 
 mention the fact that in 
 a conduit where a gate- 
 house admits access to 
 the interior, the author 
 has seen the rapidly flow- ^'^- '"-Sponge spicules. 
 
 ing water swarming with these creatures, so that, in dipping 
 up a single glass of water, a dozen or twenty of the hydras 
 would bo taken at the same time. On the bottom and sides 
 
ANIMAL AND VEGETABLE LIFE. 
 
 83 
 
 Fig. II. — Hydra viridis. 
 
 of the conduit were, no doubt, hundreds of thousands of the 
 
 little animals, but no recogniz- 
 able effect was produced on the 
 water by their presence or by 
 their death and decay, which 
 must have followed. 
 
 Of course there are many 
 other animal inhabitants of fresh 
 waters, and when we descend to 
 microscopic organisms we have 
 a very great variety of forms. 
 The term Infusoria is often used 
 to cover all these minute animal 
 
 organisms, which are grouped under orders and families and 
 
 genera and species Hke the higher 
 
 animals. Scarcely any natural 
 
 water is free from these infusoria, 
 
 and some of them are peculiar to 
 
 impure waters, but, as they can 
 
 be recognized and studied only 
 
 under the microscope by those familiar with 
 
 such matters, we need not dwell upon them 
 
 here. 
 
 Plants. — Generally speaking, the flowering 
 
 aquatic plants, such as are known as eel-grass, 
 
 pond-weed, pickerel-weed, etc., are in them- 
 selves of no disadvantage, while growing, to the 
 
 pond or reservoir in which they grow ; they are, 
 
 indeed, of positive advantage, in oxygenating, 
 
 and so, to a certain extent, purifying the water. 
 
 If such plants, or portions of them, decay in a 
 
 limited volume of water they produce a very of- 
 fensive smell and taste. This is, in general, true 
 
 of plants aquatic and non-aquatic. It is well 
 
 known that the water in which flax is " retted " 
 
 becomes very offensive.* In some experi- 
 ments made by the author a few years ago, 
 
 the worst smell obtained was from the seed- 
 
 FlG. 12. — Pond-weed 
 (Potamogeton). 
 
 * In this connection see Reichardt, E., " Schadliche Wirkung des Rostwassers 
 von Flachs und Hanf fur die Fisclizucht." Arch. d. Pharm., ccxix, Heft i, 1881. 
 
84 WATER SUPPLY. 
 
 bearing portions of a species of Potamogeton (a common water- 
 plant), and Professor Brewer, who has made much more exten- 
 sive experiments, obtained a very fishy odor from the decay in 
 water of the leaf-stalks of a pickerel-weed, Pontederia cordata, 
 which grows on the margins of the pond from which New Haven 
 receives its supply (Whitney Lake). 
 
 While the odors and tastes obtained from different plants 
 differ from each other, in a stream or pond where the volume 
 of water is comparatively large and the opportunity for aeration 
 is great, the various tastes seem to ble7id into a more or less 
 marked marshy or pondy flavor. Sometimes, even in large bodies 
 of water, a distinctive taste is noticed ; thus, in the fall of the 
 year, the water of our ponds and lakes which are surrounded by 
 woods acquires more of a bitter or astringent taste,- which 
 is to be referred to the dead leaves at that season most 
 abundant. 
 
 A word or two may be in place with reference to the action 
 of fresh water upon vegetable matter in its bearing upon im- 
 pounding reservoirs. When vegetable matter decays in moist 
 soil, it is converted into a brown or black substance generally 
 known as humus; this is really a mixture of a number of differ- 
 ent bodies, and from it chemists have isolated a variety of sub- 
 stances, such as humic acid and humin, ulmic acid and ulmin.* 
 The acids of the humus, by oxidation, undergo chemical change, 
 to be sure, being converted into crenic and apocrenic acids, 
 which, or rather the salts of which, are found in surface waters ; 
 but when the vegetable matter is 'thoroughly "humified," as in 
 the case of peat, it exerts apparently no bad effect on the water, 
 except by giving it a brown color and a somewhat earthy taste. 
 
 When a recently felled tree is exposed to the action of the 
 water, or when bushes or even grass and weeds are killed by being 
 flooded with water, the sap and more soluble matters are leached 
 out and putrefy, or, in the presence of much air, undergo other 
 forms of decomposition. This action will take place, no mat- 
 ter under what depth of water the vegetable matter may be 
 placed, but the effect will be less marked as the amount and 
 motion of the water is greater. 
 
 * For a resume of the investigations on the composition of humus, see Julian: 
 Proc. Amer. Assoc, xxviii (1879), p. 313 and foil. 
 
ANIMAL AND VEGETABLE LIFE. 85 
 
 After the more soluble portions are extracted, the subsequent 
 decay proceeds with extreme slowness, provided the remaining 
 cellulose or woody fiber is kept continually covered with water, 
 but alternate exposure to air and water soon causes decay, as 
 every one knows. In a natural or artificial reservoir the inevita- 
 ble variations of level are very disadvantageous. As the level is 
 lowered, those aquatic plants which grow in shallow water die, 
 and if the water rises after only a short interval it becomes 
 impregnated with the products of their decay ; if a considerable 
 interval elapses, land plants grow upon the exposed surface, and, 
 being drowned by the rising waters, tend to its contamination in 
 the same manner. 
 
 It appears from this, that in the construction of impounding 
 reservoirs, the mass of growing plants, as well as the soil in 
 which they have their roots, and which of itself contains more or 
 less soluble organic matter, should be removed as thoroughly as 
 possible, especially if the water is to be of no great depth above 
 it when the reservoir is flooded. If the reservoir is filled with- 
 out such removal of the organic accumulations, a long time may 
 may be required before the chemical changes have completed 
 themselves and the water become well suited for use, but the 
 complete removal of the soil, that is, as' far as such removal is 
 practicable, is not a guaranty that no trouble will arise from a 
 newly filled reservoir. Occasionally the vegetable decay in a 
 new reservoir gives rise to much offence from the formation of 
 sulphuretted hydrogen. A marked instance of this occurred 
 in one of the basins of the Sudbury River supply, Boston, 
 Mass., the summer after it was first filled. The whole mass of 
 water in the basin was permeated with the odor, which was so 
 strong on the leeward side of the pond as to incommode the 
 passers-by. The odor was not that of pure sulphuretted hydro- 
 gen as prepared in the laboratory, and the gas was no doubt ac- 
 companied by other chemical products. The water drawn from 
 the depths of the pond had the odor of an antiquated privy. 
 The presence of sulphuretted hydrogen was made very manifest 
 by suspending in the gate-house cloths wet with a solution of 
 acetate of lead ; these became yellowish-red, and finally jet black, 
 owing to the formation of sulphide of lead. 
 
 The formation of the sulphuretted hydrogen is readily ex- 
 plained. The flooding of the basin started the decay of a large 
 
86 
 
 WATER SUPPLY. 
 
 quantity of organic matter; this taking place in the presence of 
 the sulphaies contained in the water changed them into sulph- 
 zdes, and from these sulphides thus formed sulphuretted hydrogen 
 is liberated by the acid products of decay. This same change 
 takes place to a less degree in almost all ponds and reservoirs. 
 The gas is formed, however, mainly at the bottom, and as it dif- 
 fuses upwards and mixes with the overlying water it comes into 
 contact with the oxygen in the water and is decomposed. The 
 sulphur is set free and sinks to the bottom, or in a very finely 
 divided state flows off with the water. In salt or brackish water 
 which receives sewage, these changes take place on a much 
 greater, scale. A while ago the author had occasion to examine 
 a number of samples of mud from the lower part of the Charles 
 River, a tidal stream which receives a portion of the sewage of 
 Boston, Mass. In all cases sulphur in considerable quantity 
 could be extracted from the mud by the use of proper solvents.* 
 The plants which give the most trouble in connection with 
 water supplies belong to the class of cryptogamous (non-flower- 
 ing) plants which the botanists call a/gez — plants which grow in the 
 water, or in moist places, and usually contain chlorophyll (green 
 
 coloring matter) or some 
 allied substance. Not 
 all algse are, however, 
 harmful. The so-called 
 confervoid growths are 
 made up of plants of fila- 
 mentous structure, grass- 
 green, or in some cases 
 bluish-green in color, 
 forming tangled masses 
 readily removed from 
 the water, and, when so 
 removed, shrinking enor- 
 mously in apparent bulk, 
 and drying away to a 
 grayish or colorless mass, 
 in some cases looking al- 
 most like coarse paper. 
 
 Fig. 
 
 i 
 
 -a, Diatom {S/nurnneis) ; b, Desmid 
 
 {Euastrum) ; c, Desmid {Micrasieiias). 
 
 * Eighth Annual Report Boston City Board of Health. (1879-80), pp. 12-18. 
 
ANIMAL AND VEGETABLE LIFE. 87 
 
 Plants of this character grow in almost all reservoirs or other 
 bodies of water exposed to the light and air, both in still and 
 running water ; they either float about in masses, or are attached 
 more or less'' firmly to rocks and stones and other solid objects. 
 By their growth they do no harm to the water in which they 
 flourish ; and as they are readily arrested by ordinary wire 
 screens, or easily removed by rakes or scoop-nets, their presence 
 causes no serious inconvenience in water used for town supply. 
 
 Then there are the diatoms and desmids (Fig. 13), which are 
 interesting objects under the microscope, and occur in consider- 
 able abundance and great variety ; they are not, however, as far 
 as we know, of any significance in surface waters, at least, from a 
 sanitary point of view. 
 
 The vegetable organisms which cause the most trouble and 
 inconvenience are those which appear as greenish specks, or mi- 
 nute straight or curved threads, diffused through the water — 
 visible enough if a large quantity of water be looked at, but per- 
 haps almost escaping notice in the small quantity which would 
 be taken up in a single glass. It is true that the individual 
 plants are in some cases distinguishable by the naked eye ; but 
 their form and structure can be made out only by use of the 
 microscope. If collected together as a scum, which often hap- 
 pens, especially on the windward shore of a pond, the scum is 
 not coherent, is easily broken up, either by a wind setting in the 
 opposite direction, by a shower of rain, or by artificial agitation. 
 The appearance has been sometimes described as that of meal or 
 of fine dust scattered through the water. The number of indi- 
 viduals is almost infinite ; and under favorable conditions they 
 increase with great rapidity. Their presence gives a decidedly 
 green or greenish yellow tinge to large bodies of water ; and 
 their death and decay often cause considerable offence to the 
 sense of smell of those sojourning in the neighborhood, and to 
 the sense of taste of those obliged to drink the water. The 
 troublesome species belong, almost all, to the family of Nostocs, 
 and of these the number which have been known to produce 
 difficulty is small ; this may, of course- be due in part to imper- 
 fect observation. A single species each of Cozlosphmriiim and 
 Clathrocystis, two or more of Anabmna and one of Sphcerozyga 
 have been observed in considerable quantities in the neighbor- 
 hood of Boston, Mass. 
 
88 
 
 WATER SUPPLY. 
 
 Fig. 14, c gives a general idea of the appearance of the Cla- 
 throcysiis ceruginosa when magnified some 300 diameters. This 
 plant is often found in much larger masses than indicated in the 
 
 Fig. 14. 
 
 cut ; in fact, the little sacfk-like masses are sometimes large 
 enough to be made out by the unaided eye, although no idea of 
 the structure can be thus obtained. Fig. 14, a attempts to give 
 an idea of the Anabana circinalis, one of the Nostocliinea; ; this 
 plant occurs very frequently in ponds and in sluggish streams. 
 Another common variety of the same genus is similar, except 
 that the filaments are straight instead of curved ; and there are 
 other genera of algcB which occur in the same way as the Ana- 
 bcena, and present a similar appearance. 
 
 These alga;, when present in any considerable quantity, give 
 a repulsive appearance to the water, and when they are in a state 
 of decay they communicate to it an offensive taste and odor. 
 Fortunately, in most cases, the trouble which they cause is of 
 short duration, although often recurring in the same water sup- 
 
ANIMAL AND VEGETABLE LIFE. 89 
 
 ply year after year. Their presence is not a sign of contamina- 
 tion, as they occur in natural ponds removed from all polluting 
 influences. While, however, they do grow in pure waters and in 
 old and clean ponds, they seem to grow more abundantly in water 
 containing mud and vegetable extractive matter, as in newly 
 filled reservoirs ; so that, while immunity from their presence 
 cannot be guaranteed in the case of any pond, they may with 
 some certainty be looked for in dirty and especially in shallow 
 ponds. A warm temperature and shallow water are perhaps 
 of even more importance than the products of decay of higher 
 plants, for all surface waters contain the ammoniacal and mineral 
 salts necessary for the growth of the algs. 
 
 Whether the presence of these minute algae gives an unwhole- 
 some character to a water which is otherwise suited for domestic 
 use, is an open question ; but such information as the author has 
 been able to get from various sources coincides with the state- 
 ment of the Mass. State Board of Health, who investigated the 
 matter when a certain portion of the water supply of Boston was 
 affected in this way. They say * that the evidence " tends to 
 show that the plant acts mechanically chiefly, perhaps like unripe 
 fruit, when affecting the health at all, in causing diarrhoea ; but 
 that the filtered water is harmless." It is known that fish often 
 die in ponds containing an abundance of the scum-forming algse, 
 probably on account of the cutting off of the supply of air ; there 
 is also one case on record where cattle have been killed by drink- 
 ing pond water which contained large quantities of a species of 
 Nodularia, a plant which has something of a resemblance to the 
 Anabcsna.\ This was in Australia. No such cases have come to 
 the knowledge of the writer here. When the algae are alive and 
 fresh, horses and cattle drink the water readily, in preference to 
 spring water : when decay takes place, the water sometimes 
 becomes so offensive that they refuse to drink it. In this condi- 
 tion it is manifestly unsuited for domestic use. 
 
 As far as our present knowledge extends, there is nothing 
 that can be done to exterminate the algse from ponds in which 
 they occur. Sometimes in reservoirs, when the algse have col- 
 lected as a scum, it is possible to float them off from the surface 
 
 * First Annual Report of the State Board of Health, Lunacy and Charity, 1879. 
 Supplement, p. xi. 
 
 f Nature, xviii (1878), p. 11. 
 
go WATER SUPPLY. 
 
 by means of properly arranged waste pipes, and in reservoirs the 
 conditions favorable for their collection and decay may be re- 
 moved by grubbing up the lilies and other pond-weeds around the 
 borders. Special devices sometimes avail in special cases. The 
 experience at Poughkeepsie, N. Y., is instructive. Here the 
 Hudson River water was pumped on to filter-beds, thence, after 
 filtration, into a small uncovered reservoir. In summer, after the 
 temperature of the water reached 70° F., an alga, one of the 
 oscillariacecB, developed in the shallow water on the beds and in 
 the reservoir, and by its death and decay in the pipes caused much 
 trouble. The trouble occurred every summer, until the following 
 method of procedure was adopted by Mr. The. W. Davis, the 
 then superintendent. As soon as the temperature of the river 
 water approached 70°, careful watch was kept on the temperature 
 and on the quality of the water delivered. As soon as the taste 
 or odor was noticed in the city, the reservoir was shut off and 
 the water pumped directly from the river into the mains. In 
 this way all trouble was avoided and there were no complaints. 
 
 With reference to the minute organisms, animal and vegetable, 
 it is a curious fact that certain forms will sometimes suddenly 
 appear in places where for years previous they have riever been 
 known to occur, and they may disappear as suddenly as they 
 came. In other cases, forms which have been known to be pres- 
 ent to a limited extent will increase enormously, owing to con- 
 ditions of which we are quite ignorant. 
 
 Odors and Tastes of Surface Waters. 
 
 Surface waters often possess peculiar odors and tastes. These 
 are sometimes explicable, as in the case of the odor of sulphu- 
 retted hydrogen referred to on page 85. Then there are other 
 odors (with accompanying tastes), which are quite certainly due 
 to the algae. These odors are q'uite various. Mr. Fteley (Sud- 
 bury River) speaks of a musty odor ; at Albany it was spoken of 
 as a musty and cucumber odor ; at Springfield (Ludlow Reservoir), 
 the first summer after the reservoir was filled there was a most 
 distinct odor of green corn, perceptible for a quarter of a mile 
 from the pond on the leeward side. The pond was covered with 
 a slime of algae, partially decayed ; and the same marked odor 
 was noticed at the water troughs along the line of the aqueduct. 
 
TEMPERATURE OF SURFACE WATERS. 91 
 
 When, under the excessive heat of summer, the alg^ are col- 
 lected in masses and begin to decay, a most abominable pig-pen 
 or horse-pond odor is sometimes noticed in the ponds ; but this is 
 seldom noticed in the water drawn from the service pipes, al- 
 though a foul odor similar to that comrnon in " dead-ends" does 
 occur when water containing the algae stagnates in the pipes. 
 
 Besides the tastes and odors which may with reasonable cer- 
 tainty be ascribed to the growth or decay of organized beings, 
 there have been certain conditions of the water in the case of 
 many water supplies which are very enigmatical, and for which 
 no satisfactory explanation has been offered. The so-called 
 " cucumber " taste, and the other tastes characterized as " fishy " 
 or " oily," occur when the water is of its ordinary purity and in 
 waters naturally pure. Some have maintained that the tastes 
 ought to be due to the presence or to some peculiar condition of 
 the algse, but as it is impossible to discover any unusual amount 
 or condition of those algse, which are, so far as we know, harm- 
 less, and which are always present, and as none of those algae 
 which are known to produce bad tastes and odors are found, it is 
 rather difficult to accept this explanation. Since Professor Rem- 
 sen has found reason to believe that the recent condition of the 
 water in Farm Pond is due in part, at any rate, to the decay of 
 a sponge, it has been suggested that we have here the cause of 
 the various difficulties heretofore unexplained. While it seems 
 to be undoubtedly true that the sponge may, under certain cir- 
 cumstances, produce what is properly spoken of as a " cucumber " 
 taste, there are many cases on record where it is difficult to believe 
 that this can be the cause ; and with reference to such cases, for 
 the present, we can only say, " We do not know." 
 
 Temperature of Surface Waters. 
 
 One of the great disadvantages to which surface water is 
 subject is the variation in temperature. The water in winter is 
 but a few degrees above the freezing point, and in summer the 
 temperature is so high that the water is not agreeable as a bever- 
 age unless artificially cooled. The use of ice is so general in the 
 United States that much less stress is laid upon this point than 
 in other countries, but, of course, a considerable proportion of the 
 inhabitants in the thickly settled parts of our cities are unable to 
 
92 WATER SUPPLY. 
 
 supply themselves regularly with ice. Of late years iced water 
 has been supplied at several public fountains in New York, and 
 perhaps in other cities. It would be difficult to isolate the effect 
 of the temperature of the water on the public health from the 
 general effect of the hot weather, of eating unripe and decayed 
 fruit, and of other causes, all of wliich affect particularly the 
 poorer class of the population. In England, the increased death- 
 rate of the warmer months has been connected directly with the 
 temperature of the water supply. Thus, in the report of the 
 Registrar General for July 22, 1878, we find: 
 
 " The high mortality of the week is due to diarrhoea, which 
 becomes fatal in London when the temperature of the Thames 
 rises above 60° F. Thus the Thames temperature, which had 
 been 60°, rose in the last week of June to 65°; in the following 
 weeks it was 68°, 66°, 67° F. The weekly deaths from diarrhoea 
 and simple cholera, which had been 23, rose to 78, 156, 256, 349 
 in corresponding weeks." To show that this increase was not 
 due simply to the increased atmospheric temperature and its 
 attendant discomforts, it is stated that " the deaths from diar- 
 rhoea are differently distributed in the fields of the water com- 
 panies ; thus the deaths in the last four weeks were 786 in the 
 districts supplied by the Thames and Lea waters; whereas the 
 deaths in the districts supplied with water drawn from the chalk 
 by the Kent company were 19 ; out of the same population, the 
 deaths in the former were to the deaths in the latter as 3 to i." 
 The temperature of the Kent company's water at the wells was 
 uniformly 52° F.; at the same time it must be noted that the 
 waters, differed not simply with respect to temperature. The 
 water of the Kent company is harder, and, what is perhaps more 
 important, it contains but little organic matter. 
 
 Baldwin Latham, while believing that " the summer diarrhcEa 
 is governed by the influence of the temperature of our water 
 supply, as invariably the disease becomes epidemic when the 
 water, whatever be its source of supply, reaches a temperature 
 of 62° F. (i6°.7C.)'," attempts to show that the temperature of 
 the water delivered to the consumers is much less dependent 
 upon the original temperature at the source than is usually sup- 
 posed, and that if the water is carried for any considerable dis- 
 tance in the mains, it approaches or acquires the temperature of 
 the ground in which the pipes are laid, as appears from the 
 
TEMPERATURE OF SURFACE WATERS. 
 
 93 
 
 following table.* Here the Kent water, alluded to as having a 
 uniform temperature at its source, appears, when delivered, as 
 variable as the water of the Thames. He asserts f that the 
 general mortality in London from diarrhoea is practically the 
 same in the districts supplied from the rivers as in those supplied 
 by the Kent Water Company, but that the water in the latter 
 region does not reach its highest temperature until later in the 
 season, owing to the advantage which the lower initial tempera- 
 
 TABLE X. — Temperature of London Water Supplies. 
 [The degrees are Centigrade.] 
 
 Date. 
 
 July, 1876 
 
 August, 
 September, 
 October, 
 November, 
 December, 
 January, 1879 
 February, 
 March, 
 April, 
 May, 
 'June, 
 
 Kent Company's Water. 
 In the wells. In the mains. 
 
 10.59 
 10.67 
 10.71 
 10.68 
 10.61 
 10.50 
 10.45 
 10.40 
 
 10.35 
 10.44 
 10.42 
 10.57 
 
 16.80 
 
 16.62 
 
 14.78 
 
 12.90 
 
 8.17 
 
 5.98 
 
 5.06 
 
 5-45 
 6.30 
 8.38 
 9.36 
 12.63 
 
 Thames 
 Water in 
 the Mains. 
 
 18.39 
 
 17.70 
 
 14.92 
 
 12.87 
 
 7-47 
 
 5.14 
 
 4-77 
 
 5-72 
 
 7.20 
 
 8.20 
 
 10.20 
 
 14.25 
 
 The Ground at a depth 
 OF meters 
 
 1.448. 
 
 17-66 
 
 15.47 
 
 17.16 
 
 16.05 
 
 15.28 
 
 15.04 
 
 12.53 
 
 13-05 
 
 6. 98 
 
 904 
 
 3-72 
 
 6.12 
 
 2.80 
 
 4.68 
 
 3.58 
 
 4.66 
 
 • 4-97 
 
 5-54 
 
 6.80 
 
 6.85 
 
 7.32 
 
 8.31 
 
 13.01 
 
 11.23 
 
 ture gives to it. It will be understood that Latham's figures have 
 reference to the temperature acquired in the service pipes;:} the 
 water passing through the main conduit or even through the 
 large iron mains suffers much less change of temperature than 
 Table X would indicate.§ Latham has patented an apparatus 
 for " tempering " the water, which consists " of a vertical tube 
 driven or screwed into the ground to a depth of about 25 feet, the 
 water being admitted at the top and withdrawn at the bottom, 
 and special arrangements being adopted for the protection of the 
 ascending pipe." With such an apparatus interposed in the 
 service connection of the house, " the range of temperature in 
 
 ♦Quoted from Journal flir Gasbeleuchtung und Wasserversorgung, xxii (1879), 
 P- 756- 
 
 f Journal of Society of Arts, Sept. 17, 1880. 
 
 J The temperatures given in the reports of Dr . Frankland to the Registrar Gen- 
 eral were taken in the company's mains at Deptford, near the source of the water. 
 
 § This matter is discussed at length and mathematically by Perissini : Jouni. fiu: 
 Gasb. und Wasserv., xxiii (1880), pp. 608 and 644. 
 
94 
 
 WATER SUPPLY. 
 
 the water required for dietetic purposes need not exceed 3° F. 
 throughout the year, when drawn from a 3-inch tube at a rate 
 not exceeding one gallon every half-hour." * 
 
 The question of temperature is an important one in bodies of 
 stored water, as the troublesome algae already alluded to seem to 
 require a somewhat elevated temperature (approaching 70° F.) 
 for their rapid and abundant development ; when collected as a 
 scum they are killed and enter into decay when the water be- 
 comes strongly heated in midsummer. While it is, of course, 
 impracticable to cover ponds and impounding reservoirs, it is of 
 advantage to cover the smaller storage reservoirs which are fre- 
 quently used in connection with water-works to contain a reserve 
 supply, or a supply for a limited number of days, and thus to pre- 
 vent, to a certain extent, the elevation of temperature to which 
 
 TABLE XI. — Observations of Temperature in Fresh Pond, Mass.* 
 
 DaA. 
 
 Temperature (expressed in Centigrade 
 degrees) at 
 
 U CO u 
 
 u u u 
 
 May 4, 1878, 
 
 14, 
 
 June 4, 
 
 " 12, 
 
 19, 
 
 25: 
 
 July 2, 
 
 9. 
 
 16, 
 
 " 23, 
 
 August 6, 
 
 13, 
 
 20, 
 
 27, 
 November 7i. 
 December 7, 
 January 2, 1879 
 
 14, 
 
 " 22, 
 
 April . 14, 
 
 May 13, 
 
 * First Annual Report of Mass. State Board of Health, Lunacy and Charity, 1880, 
 Supplement, p. 98. 
 
TEMPERATURE OF SURFACE WATERS. 95 
 
 a small and comparatively shallow body of water is subject under 
 a summer's sun. 
 
 Table XI shows the variation in temperature of the surface 
 water in a pond near Boston, Mass. (lat. about 42°), which is used 
 as a source of city supply ; and also the variation at different 
 depths. 
 
 For making observations on the temperatures of bodies of 
 water, especially below the surface, thermometer makers supply 
 instruments surrounded by a copper tube in which some of 
 the water is brought to the surface. Where samples for an- 
 alysis are taken at the same time, a chemical thermometer 
 may be inserted in the bottle used for the collection of the 
 water. The bottle having been sunk to the required depth, the 
 stopper is withdrawn, and after the bottle has filled it is allowed 
 to remain until it is certain that the whole apparatus has acquired 
 the actual temperature of the water. It is then drawn up and 
 the reading of the thermometer is taken quickly while still sur- 
 rounded by water. This method answers very well, with care, up 
 to depths of 80 feet. 
 
 A very convenient instrument for taking temperature at va- 
 rious depths is the " New Standard Deep Sea Thermometer," 
 made by Negretti and Zambra, London. The construction of 
 the thermometer is shown in Fig. 15, t. To protect it against 
 pressure, this thermometer is inclosed in a glass tube, hermet- 
 ically sealed, the portion which surrounds the thermometer 
 bulb being filled with mercury. The object of the mercury is to 
 furnish a good conductor of heat between the outer wall and the 
 thermometer bulb, and it is confined in place by a partition ce- 
 mented on to the neck of the bulb. The whole apparatus is 
 inserted into a hollow wooden frame containing a quantity of 
 lead shot. When the thermometer is lowered to any depth, it 
 descends as shown in Fig. 15, a, and the bulb of the thermome- 
 ter is downward ; it is allowed to remain at the required depth 
 for a few minutes in order that the thermometer may acquire the 
 temperature of the place, the mercury rising or falling in the 
 capillary tube as in an ordinary thermometer. Finally, a sudden 
 pull is made on the line, and the instrument, owing to the resist- 
 ance of the water and the consequent displacement of the center 
 of gravity (the shot falling to the other end of the frame), will 
 ■ turn over and be drawn upward with the thermometer in the 
 
g6 
 
 WATER SUPPLY. 
 
 position shown in Fig. 15, b. When the thermometer is inverted 
 the mercury column breaks at the constriction A, and falls to the 
 other end of the tube, from which the degrees are read off, as 
 shown in the figure, with the bulb of the thermometer upper- 
 most. 
 
 As with any thermometer, it is necessary to determine the 
 
 DESCENDING. 
 
 Fig. 15, a. 
 
 ASCENDING. 
 
 Fig. 15, (. 
 
 Fig. 15, b. 
 
 error of graduation once for all, and the error of the zero- 
 point from time to time. For very nice work, a correction 
 should also be made for the temperature of the mercury column 
 at the time of reading, but in ordinary work this is not neces- 
 sary. 
 
 Examination of Surface Waters. 
 
 In choosing a surface water as a source of supply, there are 
 
EXAMINATION OF SURFACE WATERS. 97 
 
 certain concessions which must be made. In the first place, the 
 water will be, almost inevitably, somewhat colored, especially if 
 taken from a pond or lake ; in the second place, there will usually 
 be a slight " pondy " taste, even in the best of surface waters. 
 Considered simply as drinking water, surface water will always 
 be at a disadvantage by the side of a pure, soft spring water, but, 
 as stated on previous pages, a surface water may often be, on the 
 whole, the best suited for a general supply. 
 
 In examining as to the suitability of any source of surface 
 water, after determining whether a sufficient amount can be ob- 
 tained directly or by means of storage reservoirs, the desirability 
 of the source can be ascertained better by a survey of the drain- 
 age area, and by a knowledge of the present population and 
 sources of pollution, and the probable increase in the future, than 
 from the results of chemical examination of the water, the inter- 
 pretation of which is sometimes attended with difficulty. It 
 may be possible, it is true, to state from the chemical examina- 
 tion of a single sample that no considerable or no appreciable 
 contamination exists ; it is impossible to recommend a water for 
 drinking without knowing something of the situation and sur- 
 roundings of the source from which it is taken. 
 
 The principal difficulties in the way of the satisfactory chemi- 
 cal examination of surface waters are three in number: In the 
 first place, the volume of water is generally so large that, even 
 when polluting matter is known to be present, the dilution is so 
 great as to prevent the detection of unmistakable evidence of 
 contamination. In the second place, all surface waters contain 
 more or less of organic substances — substances containing carbon 
 and nitrogen — which it is -impossible to refer definitely to animal 
 or vegetable sources, or otherwise to distinguish as harmless and 
 harmful. In the third place, the water of such streams and 
 ponds is subject to very considerable variation, so that the ex- 
 amination of a single sample is of comparatively little value. 
 With reference to the second point something has already been 
 said in the chapter on water analysis. It may be said further 
 that the substances which form the most offensive part of the 
 soluble vegetable matter are albuminous in character, and the 
 chemical effect on the water is to increase the amount of what 
 is designated as " albuminoid ammonia ; " that is, they contain 
 nitrogen, which, under the analytical treatment, is evolved and 
 7 
 
98 WATER SUPPLY. 
 
 measured as ammonia. It is unfortunately impossible by ana- 
 lytical means to distinguish whether this " albuminoid ammonia" 
 is to be ascribed, in any given case, to vegetable or to animal 
 origin. No doubt the excrement of fishes, their dead bodies so 
 far as they are not consumed by their living comrades and by 
 the animalcules, and the bodies of the animalcules themselves 
 add to the nitrogenous organic matter in our surface waters. As 
 a rule, in waters not contaminated by sewage, the animal matter 
 forms only a trifling proportion of the entire organic matter, but 
 the recent investigation of Professor Remsen shows that in some 
 instances the animal matter (as from sponges) may be apprecia- 
 ble and of practical importance. 
 
 In the case of a bad condition of the water arising from the 
 decay of an unusual amount of animal or vegetable matter on 
 the bottom or sides of a reservoir, or from the presence of algae, 
 the water is ordinarily characterized by an abnormal amount of 
 soluble organic matter, which shows itself in the common method 
 of analysis as " albuminoid ammonia," or in the Frankland 
 method as " organic carbon " and " organic nitrogen." But, in 
 order to judge whether the amount is abnormal or not, it is nec- 
 essary to have an extended series of analyses, and this is seldom 
 at hand. The effect of algae may be well seen by a study of the 
 weekly analyses of the Springfield water for the years 1876 and 
 1877, and of the Mystic during 1879, ^^ shown in the respective 
 water reports.* 
 
 As, in this country, waters have been and are most frequently 
 examined by the " ammonia " process, attempts have been 
 made to fix the limit of " albuminoid ammonia " which may be 
 allowed in a surface water. Wanklyn, as stated on page 41, 
 looks upon o.oio part of "albuminoid ammonia" in 100,000 as 
 suspicious, and upon 0.015 part as sufficient to condemn a water, 
 and some analysts are inclined to adopt this standard for our sur- 
 face waters ; such an absolute standard is, however, impractica- 
 ble, and would exclude many waters which are known to be free 
 from contamination and to be perfectly well suited for domestic 
 use. With reference to this matter Dr. Smart says : + 
 
 * See Annual Reports of the Water Commissioners of the City of Spring6eld 
 (Mass.), 1877 and 1878. Also, First Annual Report of Massachusetts State Board of 
 Health, Lunacy and Charty. Supp., pages 119, 120. 
 
 •[■Buck's Hygiene, ii, p. 128. 
 
EXAMINATION OF SURFACE WATERS. 9< 
 
 " The waters of the purest mountain streams in our unsettlec 
 West, where animal contamination is an impossibility, contaii 
 0.014 part per 100,000 of albuminoid ammonia. At other time 
 they may yield 0.020, 0.025 or more, and yet be regarded as com 
 paratively innocent." 
 
 Dr. Smart found the Black's Fork, Wyoming, to contain o.oi/ 
 part in 100,000 when most free from surface admixture, an( 
 0.028 part when swollen by melting snows. The Little Wine 
 River, Wyoming, a stream running over a rocky bed and contain 
 ing only about 3.75 parts of total solids in 100,000, gave 0.03- 
 part of " albuminoid ammonia." The North Platte River yield 
 from 0.030 to 0.050 part in 100,000 at different times. The quej 
 tion may be pertinently asked how it is that our surface water 
 contain so much more organic matter than the waters of Grea 
 Britain ? Dr. Smart says : " These streams ought to be pure, i 
 pure water is to be found in nature, as they run through no pop 
 ulous districts and are thus free from the sources of contamina 
 tion against which sanitary officers are most on guard. Thei 
 spring from a cleft in the rocks, are mostly rapid in their coursi 
 until they reach the plains, and are fed by the rainfalls and melt 
 ing snows. There seems nothing left by way of explanatioi 
 but the wildness of the country through which they run. Ii 
 England the fields are fenced in and the soil cultivated to th( 
 very banks of the stream, the woods are well kept, and thi 
 swamps are drained and reclaimed ; but here there is no cultiva 
 tion ; vegetation lives, and, instead of being garnered up, die 
 and decays. The forest trees fall and rot where they fall. . 
 have been up in the Uintah Mountains, where are the sources o 
 the Black's Fork, and among the pines covering the slopes o 
 the ridge there are more fallen trees in all stages of decay thai 
 living ones in those untouched forests. Through such deac 
 vegetation the streams have to force their way, and it would bi 
 singular indeed if they did not take up a portion of the solubl( 
 organic matter. But, in addition, in the tangled willow growth; 
 of the valleys, where, as in the forests, the growth of to-day rise 
 from the decay of ages, the beaver dams up the stream, anc 
 vast masses of water are stagnated, to dissolve the dead vegeta 
 ble tissues and find their way by slow degrees back into the bed: 
 
ICX) WATER SUPPLY. 
 
 is also shown by the absence of chlorides and nitrites, and that 
 it is recent, by the frequent absence of ammonia." 
 
 It must not be understood that the presence of an excess of 
 vegetable matter is absolutely a matter of indifference. Waters 
 thus charged may be the cause of diseases of the digestive or- 
 gans, without doubt. Whether they may also cause any of the 
 so-called zymotic diseases or give rise to specific diseases of pe- 
 culiar type, is doubtful. 
 
 Dr. Smart, whose opinion is entitled to much weight, ascribes 
 the " mountain fever " of the West to this caiise (compare page 
 52). He says: 
 
 " As animal matter, in certain stages of its decomposition, or 
 enveloping specific germs, gives expression to its presence in the 
 system by the development of typhoid fever, so vegetable mat- 
 ter, in certain stages of its decomposition, or enveloping specific 
 germs, gives rise to an adynamic remittent, for which the writer 
 has suggested the name of aquamalarial fever. Viewed in its 
 connection with this affection, the organic, ammonia should not 
 exceed 0.016 part in 100,000, for when 0.020 is reached, the disease 
 makes its appearance, and becomes more pernicious in individual 
 cases as the amount increases. Not that aquamalarial develop- 
 ments are to be expected in every case where the organic ammo- 
 nia exceeds this limit, but that experience shows a greater prob- 
 ability of their occurrence with a water thus impregnated — ma- 
 laria being more likely to be present with a large than a small 
 vegetable contamination." 
 
 With reference to the third point alluded to above as afford- 
 ing difficulty in the interpretation of the results of chemical ex- 
 amination, it may be said that the results of a single examination 
 are to be received with a good deal of caution, and such results 
 must be interpreted somewhat according to the season of the 
 year in which the sample is taken. The very considerable varia- 
 tion to which streams are subject, as far as the suspended matter 
 is concerned, has been shown in Table VII, on page 57. The 
 state of things is similar with reference to the matter in solution, 
 especially with streams in which the volume of water is subject 
 to much variation. Thus the Hooghly, at Calcutta, is stated 
 to carry : 
 
 21 .68 parts in loo.cxjoof "total solids" in May, 
 11.30 " " " " " " " October. 
 
EXAMINATION OF SURFACE WATERS. 
 
 lOI 
 
 The rainy season greatly increases its purity, and the melting 
 of the snow on the Himalayan Range helps in the same direction, 
 so that the water is most pure in October.* Further illustration 
 of the variation to which surface waters are subject, particularly 
 with regard to the organic matter, may be seen from the follow- 
 ing tables. Table XII shows the variation in the sum of the 
 " organic carbon " and " organic nitrogen," and also in the " albu- 
 minoid ammonia," as observed in the case of the Mystic water, 
 which is sup_plied to a portion of Boston, Mass. 
 
 TABLE XII. — Examination of Mystic Water. 
 [Results expressed in Parts in icx),cx)o.] 
 
 Date. 
 
 No. OF Samples. 
 
 Sum of the 
 
 Organic 
 Elements. 
 
 " Albuminoid 
 Ammonia." 
 
 
 Average of 2 samples 
 " 4 " 
 
 " 4 " 
 
 " 3 " 
 
 " 5 
 " " 2 " 
 " " I " 
 
 " 2 " 
 
 " 3 " 
 
 " 4 " 
 
 " 4 " 
 
 " " 12 " 
 
 0.478 
 
 0.775 
 0.605 
 0.448 
 0.333 
 0.393 
 0.297 
 0.364 
 0.486 
 0.852 
 0.526 
 0.375 
 
 
 Tulv 
 
 0.035 
 0.026 
 
 
 September 
 
 
 0.014 
 
 November 
 
 
 
 January 1880 
 
 
 
 
 July 10 to Aug. 7 
 
 Aug. 14 to Sept. II 
 
 Sept. 18 to Jan. 15 
 
 0.034 
 0.022 
 0.014 
 
 In this case, the large variation was partly due to the pres- 
 ence of an abundant growth of algs, the trouble being at its 
 height in the latter part of July and during August. But, in the 
 absence of any such growth, the variation is often very great. 
 Weekly examinations of the Cochituate supply of the city of 
 Boston from July, 1876, to July, 1877-, showed a variation 
 
 in the ammonia from 0.0005 to 0.0056 part in 100,000 
 
 in the " albuminoid ammonia " from o.oogg to 0.0176 " " 
 
 in the total solids from 3.72 to 5.58 " " " 
 
 Further illustration of this point is afforded by Tables XIII 
 and XIV. 
 
 Table XIII contains the results of monthly examinations of 
 the water of Glasgow, Scotland, as reported by Professor Gustav 
 Bischof of the Andersonian University. The water comes from 
 
 *Proc. Inst. Civ. Eng. Gr. Br., Ixiv, p. 361. 
 
I02 
 
 WATER SUPPLY. 
 
 TABLE XIII.— ExAMiNATioK OF Glasgow Water by Professor Bischof. 
 
 [Results expressed in Parts in 100,000.] 
 
 Date. 
 
 May 12, 1873 
 
 June 16, 
 
 July 18, 
 
 Aug. I, 
 
 Sept. I, 
 
 Oct. 14, 
 
 Nov. 24, 
 
 Jan. 15, 1874 
 
 Feb. 5, 
 March 10, 
 
 April I, 
 
 May 5, 
 
 Average . . 
 
 Organic 
 
 Carbon.- 
 
 0.204 
 0.181 
 0.192 
 0.209 
 0.156 
 0.256 
 0.177 
 0.154 
 0.251 
 0.185 
 0.169 
 0.235 
 
 0.197 
 
 Organic 
 
 Nitrogen. 
 
 0.017 
 0.016 
 0.008 
 O.OII 
 
 0.047 
 
 0.028 
 
 0.035 
 
 U.02I 
 0.008 
 0.007 
 0.003 
 O.OII 
 
 0.018 
 
 Sum of Or- 
 ganic Ele- 
 ments. 
 
 0.221 
 0.197 
 0.200 
 0.220 
 0.203 
 0.284 
 0.212 
 0.175 
 0.259 
 0.192 
 0.172 
 0.246 
 
 0.215 
 
 an unpolluted Highland lake, Loch Katrine, and flows some 
 thirty-six miles in a masonry conduit. Table Xiy, which, like 
 No..>XIII, is compiled from the Sixth Report of the Rivers Pol- 
 lution Commission, shows the variation in the water of the sev- 
 eral companies which supply London with filtered river water. 
 These are not monthly averages, but the results of the examina- 
 tion of single samples taken at monthly intervals. 
 
 TABLE XIV. — Variation in Monthly Samples of London Water, 1873. 
 
 Name of Company. 
 
 Maximum 
 
 at any one 
 
 time. 
 
 Chelsea 
 
 West Middlesex 
 
 South wark 
 
 Grand Junction. 
 
 Lambeth 
 
 New River 
 
 East London. . . 
 
 Organic Carbon. 
 
 Minimum 
 
 at any one 
 
 time. 
 
 0.447 
 
 0.341 
 
 0.396 
 
 0.412 • 
 
 0.449 
 
 0.257 
 
 0.333 
 
 O.I2I 
 
 o. 114 
 
 O.II8 
 O.II7 
 0.130 
 0.059 
 0.109 
 
 Mean 
 
 of 12 
 
 Samples. 
 
 0.197 
 0.173 
 0.186 
 0.183 
 0,206 
 0.107 
 0.175 
 
 Organic Nitrogen. 
 
 Maximum 
 
 Minimum 
 
 Mean 
 
 at any one at any one 
 
 of 13 
 
 time. 
 
 time. 
 
 Samples. 
 
 0.067 
 
 0.013 
 
 0.034 
 
 0.055 
 
 0.015 
 
 0.028 
 
 0.060 
 
 0.020 
 
 0.030 
 
 0.050 
 
 0.016 
 
 0.032 
 
 0.065 
 
 0.021 
 
 0.040 
 
 0.032 
 
 O.OIO 
 
 0.018 
 
 0.082 
 
 0.015 
 
 0.035 
 
 It may be said, in general, that no source of surface supply 
 should be adopted on the strength of examinations made in the 
 winter season, and in the case of ponds and lakes the examina- 
 tion should include a careful survey of the shallow portions 
 during the hot weather, with a view of discovering the possible 
 presence of noxious algae. Negative results in this direction will 
 not, however, guarantee future immunity. 
 
EXAMINATION OF SURFACE WATERS. 
 
 103 
 
 As already intimated, the preliminary examination of a sur- 
 face water, from a sanitary point of view, concerns itself chiefly 
 with the present and probable future pollution, and this must be 
 considered with reference to the volume of water with which the 
 polluting substances will be diluted, taking care to err rather on 
 the side of safety : the chemical examination should not, how- 
 ever, be omitted, as the results, if properly interpreted, may be 
 of great value. Tables XV and XVI will furnish data for com- 
 parison in the case of surface waters. 
 
 TABLE XV. — Examination of Surface Waters. 
 
 [Results expressed in Parts in ioo,oco.] 
 
 
 
 
 i 
 
 
 ZlS 
 
 Z 
 
 
 
 Haro.ness, 
 
 
 
 i 
 
 1 
 
 
 
 
 h 
 
 
 ^g 
 
 Q 
 
 
 
 
 
 i 
 
 
 
 
 
 a. 
 
 Geological For- 
 
 
 
 tn 
 
 < 
 
 
 
 6 
 
 ^ 
 
 
 < 
 
 w (/) H 
 
 owe 
 fc: 5 
 
 £ K OS 
 
 t: f- h 
 
 
 
 cr 
 
 u 
 c 
 
 E 
 
 
 
 mation. 
 
 
 y 
 
 < 
 
 i 
 
 s 
 s 
 
 c3§ 
 0^ 
 
 
 s 
 
 X 
 
 s 
 g. 
 
 S 
 
 
 
 (1. 
 
 
 d 
 
 
 H 
 
 
 
 
 
 < 
 
 '^ 
 
 H 
 
 PL. 
 
 u 
 
 H 
 
 a. 
 
 H 
 
 ^ 
 
 Igneous rocks ■ 
 
 Metamorphic, Cam- 
 
 5-^5 
 
 0.278 
 
 0.033 
 
 O.OOI 
 
 0,002 
 
 0.035 
 
 
 
 '■■3 
 
 O.l 
 
 2.0 
 
 2.1 
 
 18 
 
 brian, Silurian, 
 
 
 
 
 
 
 
 
 
 
 
 
 
 and Devonian... . 
 
 S.12 
 
 0.293 
 
 0.024 
 
 0.002 
 
 0.006 
 
 0.031 
 
 3 
 
 0.92 
 
 0.3 
 
 2-5 
 
 2.8 
 
 81 
 
 Yoredale and Mill- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 stone Grit, and 
 
 
 
 
 
 
 
 
 
 
 
 
 
 non-calc a r e u s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Coal-measures . . . 
 
 8.75 
 
 0.377 
 
 0-033 
 
 0.003 
 
 O.OIO 
 
 0.050 
 
 6 
 
 1.05 
 
 0.4 
 
 4-3 
 
 4-7 
 
 47 
 
 Calcareous portion 
 
 
 
 
 
 
 
 
 
 
 
 
 
 of Coal-measures, 
 
 22.79 
 
 0.346 
 
 0.033 
 
 0.007 
 
 0.016 
 
 0.056 
 
 33 
 
 i.52 
 
 4.0 
 
 8.3 
 
 12.3 
 
 26 
 
 Surface Waters 
 
 
 
 
 
 
 
 
 
 
 
 
 
 FROM Cultivated 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Land. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Non-calcareous dis- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 tricts 
 
 9-52 
 
 0.276 
 
 0.034 
 
 0.007 
 
 o.oSg 
 
 0.128 
 
 635 
 
 1.49 
 
 0.6 
 
 4-3 
 
 4.9 
 
 31 
 
 Calcareous dis- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 tricts 
 
 30. oS 
 
 0.268 
 
 0.053 
 
 0,005 
 
 0.2S7 
 
 0.314 
 
 2,306 
 
 2.24 
 
 12.4 
 
 8.2 
 
 20.6 
 
 12 
 
 In Table XV are presented the results obtained by the Rivers 
 Pollution Commission from the examination of a large number 
 of samples of surface water from various geological regions, the 
 organic matter being indicated by the determination of the " or- 
 ganic carbon " and " organic nitrogen." In Table XVI are 
 brought together the results of the examination of various Amer- 
 ican surface waters, mostly in actual use as sources of supply, 
 the organic matter being indicated by the " albuminoid ammo- 
 nia." 
 
 * Sixth Report of Rivers Pollution Commission. 
 
I04 
 
 WATER SUPPLY. 
 
 o 
 
 u 
 
 o . 
 o- 
 U 
 
 it 
 
 Nichols. 
 Stuntz. 
 Van Nuy 
 
 = 1 
 
 f^^. 
 
 P^WU. 
 
 - fi; 
 
 ucJ 
 
 ^UH 
 
 u5 
 
 T^ ; O t>.aD 
 
 d '• ci 6 6 
 
 ^00 COvO ^00 O O ^ M « N 
 66^^666606^^66 
 
 O < 
 
 Sz 
 Si 
 
 IHt-<HMMOmWMHNCn^lHl->-:r 
 
 0000000000000000 
 
 0000000000000000 
 
 U-) O ■* 
 
 M « M 
 000 
 
 
 8Q'-iOOOmwi-.i-hmmc 
 oooooooooooc 
 
 o o -^ 
 000 
 
 N c*^ en en 
 M o o o 
 00000 
 
 'oooooooooooooodd 0000' 
 
 Ho 
 en 
 
 en ■ooO'OOO^MWOOO cno O Cl 1^0 O vO O t^ 
 ^ ■o^cnNcn^^^lH\oONO 
 
 CO 
 
 OJ 
 
 CTnvO 
 
 • . CO CO 
 
 • o w "-^ ' 
 
 Vr 'T'K.oD ^ ,; « 
 
 II I *^ i_i QJ O I"" 
 
 " - " .^ S M 2- 
 
 ►^ ^" - 4J P ■" ' ^ ' 
 
 -^^^jJc* bs w ^ bi) 
 
 000 t^* 
 
 00 CO 00 r>. ( 
 
 CO CO CO CO 
 
 ^ " .2" s S " 
 
 C -r :> 2 ?? pg H 
 « « .^ w ,_ rt .^ 
 
CHAPTER VI. 
 GROUND WATER AS A SOURCE OF SUPPLY. 
 
 A PORTION of the water which falls as rain or snow sinks into 
 the earth, and where the surface deposit is gravel or other 
 porous material overlying some impervious rock, the water col- 
 lects to constitute the ground water of the locality, the water- 
 table of the engineer. The rivers which flow through such a 
 deposit, or the ponds and lakes which are situated in it, deter- 
 mine the level of the ground water in their banks ; but as we 
 recede from the banks the water level is found to rise more or 
 less regularly, according to the character of the porous stratum. 
 Although subject to fluctuation, the ground water often main- 
 tains a very uniform relative level over large areas : its height arid 
 fluctuations are important factors in the sanitary condition of 
 any locality. 
 
 The cause of the rise of the ground water as we recede from 
 the wells — a fact which has been established by numerous obser- 
 vations in this country and in Europe * — is evident : the water 
 falling upon the whole water-bearing area would naturally raise 
 the level of the ground water ; that level cannot be permanently 
 higher than the water in the natural drainage channels in their 
 immediate neighborhood, as there is a continual passage of water 
 through the porous material to the visible streams or lakes. 
 Owing to the resistance of the material through which the water 
 passes, the surface will be an inclined one, and the amount of 
 inclination will depend upon the resistance encountered. On 
 Long Island, N. Y., the inclination is quite uniform from the 
 
 * A great many profiles resulting from actual measurements, and showing the 
 inclination of the ground water, may be found in the Berlin and Munich reports, the 
 titles of which appear on pages 217-218. In the latter the profiles include the sur- 
 face levels, the level of the ground water, and the level of the underlying impervious 
 stratum, and are very instructive. See also. The Brooklyn Water-Works and Sewers. 
 New York, 1867. 
 
Io6 WATER SUPPLY. 
 
 central ridge to the ocean or to Long Island Sound, and averages 
 about seven feet to the mile. In the valley of the Sacramento 
 the general slope is about four feet in a mile. In some localities 
 the inclination is much greater than this for limited distances; 
 thus, in the neighborhood of the Taunton (Mass.), Water Works 
 there is a fall, in what seems to be a continuous water-table, of 
 about 14 feet in 1,000. Besides the slope towards a visible 
 stream, there is often a fall in the ground water in the direction 
 of the mouth of the stream, and a corresponding flow. It will 
 be understood, however, that the " flow " has reference generally 
 to a rather slow passage through the interstices of the -ground : 
 sometimes, however, on account of a lack of homogeneousness in 
 the water-bearing stratum, veins of water occur, in which the 
 flow may be quite rapid. . 
 
 The water obtained by sinking a well into a stratum of sand 
 or gravel which has not been artificially disturbed, is, as a rule, 
 bright and clear, and free, or nearly free, from organic matter. 
 Although originally coming from the atmosphere, in its slow 
 passage into and through the ground, the water has been sub- 
 jected to a long process of sedimentation and filtration, com- 
 bined with processes of oxidation. In this sense, the water may 
 be said to have been purified by natural filtration: the process, 
 however, is not brought about by the means taken to collect and 
 utilize the water, but has been practically completed before the 
 demand is made upon it. 
 
 Utilization of the Ground Water. 
 The most simple method of making the ground water availa- 
 ble, and the one earliest adopted, is to sink a well, covered or 
 open, into the water-bearing stratum, and to pump therefrom. 
 Such a well, which draws its supply from the ground water proper, 
 is generally called a shallow well, in distinction from a deep well, 
 which may extend into a rocky stratum, and obtain its supply from 
 a water-bearing fissure ; and in distinction also from an artesian 
 well sunk to a considerable depth into underlying strata which 
 have no connection with the ground water of the particular 
 locality. This method of obtaining water from shallow wells has 
 been practised from time immemorial for the supply of single 
 families and small communities, and is, nowadays, used on a large 
 scale for furnishing town supplies. In many places the well, from 
 
GROUND WATER AS A SOURCE OF SUPPLY. 
 
 107 
 
 its dimensions and shape, may more properly be called a basin, 
 and, like the ordinary circular wells, these basins may be either 
 open or covered. 
 
 ' A second method of collecting the ground water is by means 
 of a covered gallery or tunnel constructed in part of porous 
 material; often the top and sides are built of tolerably impervi- 
 ous masonry or brickwork, while the bottom is of an open char- 
 acter, so that the water which rises into the gallery shall come 
 mainly from beneath. An example of this now quite common 
 mode of collection is at Lowell, Mass.'* The " filtering gallery " 
 is situated on the northerly shore of the Merrimack River and 
 parallel with it, about 100 feet from the water's edge. Its length 
 is 1,300 feet, width 8 feet, and height (inside) 8 feet. The side 
 walls have an average thickness of 2% feet and a height of 5 feet, 
 and are constructed of heavy rubble masonry, laid water-tight in 
 hydraulic mortar. The walls support a semicircular brick arch. 
 
 
 pp^>^.m&'^^^' 
 
 ^;»r,^, ••;v"<''■'^"-V^°•■•■?■■^'fS';P^^:^°--'/■"^?'.>^-' /■■'■-'?-■' 
 
 Fig. 16. — FILTER GALLERY, LOWELL, MASS. (FROM FANNING.) 
 
 one foot thick, made water-tight. Along the bottom, stone braces, 
 one foot square and eight feet long, are placed, ten feet from 
 
 * A full description of the works is given in the Fifth Annual Report of the 
 Mass. State Board of Health. Also, in the Third Annual Report of the Water Com- 
 missioners of the City of Lowell, 1873. 
 
I08 WATER SUPPLY. 
 
 center to center, between the walls, to keep them in position. 
 The water comes in at the bottom, which is covered with screened 
 gravel and small stones. 
 
 One of the best examples of a filtering gallery in this country 
 is that which supphes the city of Columbus, Ohio. This gallery 
 is a brick and cement conduit, 36 by 42 inches, and is shown in 
 process of construction in Figure 17. It is in all 5,715 feet in 
 
 Fig. 17.— filtering gallery, columbus, ohio. 
 
 length, and is situated in the gravel deposit at the junction of 
 the Scioto and Olentangy rivers.* This method of construction 
 has been followed also at Taunton, Mass. 
 
 A third method is to substitute for the collecting gallery aline 
 of iron pipes, i.e., practically water-mains, cast with a great number 
 of narrow longitudinal slits, and laid with loose joints. These 
 pipes collect the water, and conduct it to receiving wells, from 
 which the supply is pumped. In filling the trench in which the 
 pipes are laid, the pipes are surrounded on all sides with coarse 
 material, of too large a size to fall into or through the slits, and 
 the trench is then filled with screened material of decreasing size. 
 Works of this kind are in use in various places in Germany, at 
 Dresden, Hanover, etc. At Halle, glazed clay pipes are em- 
 
 * Tenth Annual Report of the Trustees of the Water-Works, Columbus, O., 1880. 
 
GROUND WATER AS A SCJURCE OF SUPPLY. 
 
 109 
 
 ployed, 47 cm. in diameter and 2.8 m. long; the total area of 
 the slits in a single pipe is 
 equal to the area of the 
 cross section of the pipe. 
 (See Fig. 18.) 
 
 Still a fourth method 
 consists in the use of the 
 "driven well," which is de- 
 scribed on page 112 and fol- 
 lowing pages, but as wells 
 of this description are fre- 
 quently sunk through a 
 layer of clay into a stratum 
 of water below the ground , 
 water proper, they partake 
 also of the character of 
 " artesian wells," which will 
 be considered in the next 
 chapter. 
 
 10 
 
 gut 
 
 -I 
 
 Fig. 18. (from fischer.) 
 
 The Effect of Pumping upon the Ground Water. 
 
 For the purposes of this discussion, we will suppose that the 
 water-bearing deposit into which a well is sunk is perfectly homo- 
 geneous. When 
 no water is re- 
 moved, by pump- 
 ing orotherwise,* 
 the water in the 
 well stands at the 
 same level as in 
 the ground; when 
 pumping begins, 
 the water falls at 
 first rapidly, af- 
 FiG. ig. terward more 
 
 slowly, until — if the discharge of the pump be uniform — a point 
 
 * If, as is exceptionally the case, the water is delivered by gravity, the effect on 
 the ground water is essentially the same as if the water were removed by pumping. 
 
no WATER SUPPLY. 
 
 is reached where the water is supphed at exactly the same rate 
 as dehvered by the pump. Suppose, in Fig. 19, that the water 
 stood originally at a b, and that an equilibrium has been estab- 
 lished when the water in the well has fallen to c. The water 
 flows from the gravel into the well by virtue of the head or dif- 
 ference of level a c, but its flow is impeded by the resistance due 
 to interstitial friction against the particles of sand or gravel, and 
 the water surface assumes somewhat the form indicated by the 
 line c d. The line c d approaches a b until it finally coincides with 
 it ; that is, there is a point beyond which no measurable influence 
 is exerted on the natural level of the ground water, and its dis- 
 tance from the center of the well depends upon the circumstances 
 of each particular case. In certain experiments made by Sal- 
 bach in the alluvial deposit on the banks of the Elbe, above 
 Dresden, it was found that when the water in an experimental 
 well was, by pumping, kept constantly 2 . 5 meters below its nor- 
 mal level, the height of the ground water was measurably affected 
 in every direction for a distance of some 60 meters (say 200 feet). 
 In experiments made by Piefke, at Berlin, the influence of the 
 well was felt to a distance of some 300 meters in wet weather, 
 and to over 700 meters at other times ; but, while the effect could 
 be traced to that distance, it was very small at the extremity of 
 this radius. With the water in the well 2.33 meters below the 
 normal level of the ground water, there was a lowering of only 
 0.35 meter at a distance of 50 meters, of 0.17 meter at a dis- 
 tance of 100 meters; and at a distance of 260 meters the lower- 
 ing was through only o.io meter. In the Berhn locality the 
 natural movement of the ground water was very slow, and 524 
 meters from the river the water stood only 0.63 meter above 
 that in the stream. 
 
 Where the ground is homogeneous, or nearly so, and the 
 water can come from one direction as easily as from another, we 
 have what may be called a circle of influence, with the center of 
 the well for the center of the circle; but if the- deposit is not 
 homogeneous, as is often the case, the water comes more easily 
 from one direction than another, and the effect of pumping is 
 felt to unequal distances from the well. As appears from Fig. 19, 
 the effect of the flow of water into the well is to form a sort of 
 crater from which the water is drained. This is often spoken of 
 as a " cone," but it is not a cone,- strictly speaking. The exact 
 
EFFECT OF PUMPING ON GROUND WATER. Ill 
 
 form of the curve c d will depend upon circumstances, but it will 
 always lie above the straight line joining the points c and d. 
 
 Returning again to our well, if pumping ceases, the water will 
 gradually flow into the well until finally it has found its level 
 and stands as in the surrounding ground. If, on the other hand, 
 starting with the water in the well at c, a larger quantity be 
 pumped than before, the level in the well will fall, and as c sinks 
 to e, the water in the ground also falls and the surface will be 
 indicated by e f. In order that this new and larger quantity 
 should be obtained, the circle of influence must become also en- 
 larged, that is, a larger area must contribute to the supply. The 
 absolute quantity of water which can be obtained from any well 
 depends upon the amount of rainfall which finds its way into 
 the ground water of the region from which the well draws its 
 supply. If the deposit were in a basin, and no water came from 
 the rain, the effect of pumping would be to gradually exhaust 
 the water : thus, if the water in the well were kept at a constant 
 level, say at c, the quantity of water delivered in a given time 
 would grow less and less, and finally the surface of the ground 
 water would be a horizontal plane passing through c ; if, on the 
 contrary, a constant quantity were pumped, the water in the well 
 Would sink lower and lower until the bottom was reached, and 
 the further delivery of the same quantity would be impossible. 
 Practically, wells are not sunk in such a deposit, but in one 
 where the ground water is continually receiving accessions from 
 the rain or melting snow, or, in some cases, from a neighboring 
 stream. The amount which can be obtained from any deposit 
 depends finally on the amount received, and no such well, how- 
 ever great its diameter or its depth, can be inexhaustible, although 
 practically it may never be called upon to furnish more than it 
 can supply. 
 
 A good illustration of the effect of a constant draught in 
 permanently lowering the level of the ground water is afforded 
 by the well in Prospect Park, Brooklyn, N. Y. The well was 
 constructed in 1869. 
 
 The elevaLion of water table as first found, Nov., 1868, was. ..15.6 ft. above tide level. 
 
 Elevation of water table, *Iay, 1879 15-2 
 
 Elevation of water table on completion of the well, Dec, 1869.14.55 
 
 Pumping began regularly on June 5, 1870; the effect has been 
 as follows: 
 
112 WATER SUPPLY. 
 
 ■5T • Average Number of Average Elevation 
 
 "^^^ Gallons per Day. o£ Water Table. 
 
 1870 300,000 14 15 
 
 1871 • 272,000 13-03 ' 
 
 1872 437,000 10.56 
 
 1873 288,600 II. 29 
 
 1874 333,000 10.70 
 
 1875 294,000 9.83 
 
 1876 235,000 g.83 
 
 1877 252,000 9.21 
 
 It is seen here that an average draught of only 304,000 gal- 
 lons per day has lowered the water table five feet in eight years, 
 and also that the yield is diminishing, for while in 1873 a daily 
 draught of 288,000 gallons permitted the water table to rise five 
 inches, in 1876 a draught of 235,000 gallons was just equal to the 
 supply, and in 1877 a draught of 252,000 gallons again lowered 
 the water.* 
 
 In the previous discussion we have spoken of a single well of 
 
 ordinary diame- 
 ter. If a series 
 of wells be locat- 
 ed near each 
 
 other, the effect 
 will be essen- 
 tially the same. 
 Fig. 20. except the circle 
 
 of influence will become approximately an ellipse, as would be 
 the case if several wells were connected so as to form an open 
 basin or a covered gallery. In locating a series of wells, reference 
 should be had to the character of the water-bearing deposit, in 
 order that they may not be placed unnecessarily near together. 
 
 Driven Wells. 
 
 " Driven " wells, " tube " wells, or as they are sometimes 
 called abroad, " American " or " Abyssinian " wells,f are formed 
 by forcing wrought-iron (or galvanized iron) tubes, such as are 
 used for gas or water pipes, down into the stratum from which 
 
 * Croes and Howell, Newark Aqueduct Board : Report on Additional Supply, 
 1879, p. 61. 
 
 f Also called "Norton" wells, the English patent having been taken out by a 
 Mr. J. L. Norton. 
 
DRIVEN WELLS. II3 
 
 the water is to be taken. The pipes are generally from iK to 2 
 inches in diameter, and furnished at the lower end with a wrought- 
 iron or steel point ; above this point the pipes are perforated for 
 some distance with holes to admit the water. The pipes with 
 point attached may be driven with a mallet or falling weight, 
 and when the top of one tube has reached the surface of the 
 ground, a second length is attached to it with a common coup- 
 ling, and the driving continued to the desired depth. In many 
 locaUties it is better to first drive down a suitable steel-pointed 
 rod or drill until water is reached, and to insert the well-tube in 
 the hole thus made.* In the perforated points for such wells, 
 the greatest variety exists, it being stated that about 150 patents 
 have been issued for points and cognate portions of the pipe. 
 The figure represents that known as Andrews' Patent. 
 
 liHiiiWiiaaMi 
 
 Fig. 21. — DRIVEN WELL POINT. 
 
 Although pipes of larger dimensions than those mentioned 
 are sometimes driven, it is usual when a large amount of water 
 is required — as for manufacturing purposes or for town sup- 
 ply — to drive a number of wells in the same limited area, and 
 connect them to a common suction pipe leading to the pump. 
 The driven well partakes of the character of the shallow well 
 when its source of supply is the ground water, but it often par- 
 takes of the character of the artesian well, as when it is driven 
 through a layer of clay or other impervious material underlying 
 the ground water into another water-bearing stratum below. 
 
 When a driven well is forced into the ground water, and 
 water is removed by pumping, the effect is essentially the same 
 as has already been described (pages 109- 112) with an open 
 well. The driven well is valuable as a means of obtaining water 
 on account of facility of construction, but it involves no princi- 
 
 * Both methods of constructing the driven well are covered by the Reissue Let- 
 ters Patent (No. 4,372) granted to Nelson W. Green, May 9, 1871, and the validity of 
 the patent has been affirmed by several legal decisions. It is claimed, however, that 
 the patent is antedated by a U. S. Patent granted to James Suggett, March 29, 1864, 
 and by British Letters Patent granted to John Goode, Oct. 16, 1823. 
 
114 
 
 WATER SUPPLY. 
 
 pie which is new as far as bringing the water to the surface is 
 concerned. 
 
 It is asserted that a driven well differs from an ordinary well 
 in two essential respects. In the first place, the well, which is 
 always of very small diameter, is not dug or bored, but sunk with- 
 out removal of the earth * in the manner already described, either 
 directly or by first driving down an iron rod, and after its re- 
 moval, inserting the well tube ; in either case the result is a nar- 
 row well with air tight walls, fitting closely in the earth about it : 
 the tube is the well. It is further asserted that when a suction pump 
 is attached to the pipe a new element is introduced, and peculiar 
 effects are produced by "exhausting the air," or "producing a 
 vacuum," and that thus the water is drawn to the pump by a 
 force independent of gravity, and to which gravity is, in this 
 case, but auxiliary. 
 
 These claims are, however, fallacious. The diagram, Fig. 22, 
 may represent the suction pipe of a pump inserted in a narrow 
 
 open well, the 
 normal level of 
 the ground wa- 
 ter being at ab. 
 Now, although 
 with a driven 
 well much stress 
 is laid upon the 
 
 a » "air-tight" tube, 
 
 it is the tube 
 alone that is air- 
 tight, as any suc- 
 tion pipe must 
 practically b e . 
 The soil, even if 
 compacted 
 
 „ about the tube, 
 
 Fig. 22. 
 
 is not air-tight, 
 and as far as transmitting pressure goes, the air which is about 
 
 * This is Col. Green's method, but well tubes are also driven into the earth at the 
 same time that their passage is facilitated by forcing into the tube, which in this <;ase, 
 is open at the bottom, a stream of water. This water washes out and brings to the 
 surface the sand and clay from the bottom in advance of the driving. 
 
DRIVEN WELLS. II5 
 
 the tube in Figure 22, and which rests on the sdrface of the water 
 at a, is in no different condition whether the tube be in natural 
 ground, or in a dug well, as indicated in the figure. The air circu- 
 lates freely in the ground ; it responds at once to any change in the 
 barometric pressure, and at once takes the place of any water 
 which may be removed from the interstices of the soil. Again, 
 below the water-level, the water, like the air above, circulates, 
 although less freely, owing to interstitial friction, and, of course, 
 fills all the pores of the ground close up to and around the pipe. 
 The statement that the pipe is the well, is misleading. If we 
 start (Fig. 22) with the tube as a suction pipe in an open well, 
 and imagine the well to be gradually narrowed in diameter by 
 filling in around the circumference, the tube will continue to be 
 practically a suction pipe, even when the well has been finally 
 filled up — the well finally having become a hollow cylinder of 
 water, of the thickness of a mere film if the soil is very compact. 
 It is quite inconceivable that the filling of the well with gravel or 
 sand, no matter how closely compacted the sand may be, could 
 produce any other effect than that due to the increased resist- 
 ance to the passage of water in the annular space about the suc- 
 tion pipe. 
 
 To show further the fallacy of the claims alluded to, let us 
 return to Fig. 22, in which the normal level of the ground water 
 is represented at a b. If, now, a given quantity of water be taken 
 continually from the point c, in a given interval of time, the slope 
 c d, which the water surface will assume, will be precisely the 
 same whether the water is drawn in buckets, or by a tube in an 
 open well, or by a driven well, because the same amount of water 
 must reach the same point c in the same time, and starting from 
 the same original position. If the claim be true that more water 
 can be obtained by one method than by another, it follows that 
 the water must be supplied faster in one case than in the other, 
 and there are two necessary consequences : first, if the point c 
 remains unchanged, this increased rapidity of flow must result in 
 an alteration of the slope of the water surface. On the other 
 hand, if the quantities pumped in the same time are made equal, 
 then in the first case the point c will not fall as low as in the 
 second. 
 
 While we should reject, a priori, the consequences to which 
 these assumptions necessarily lead, recent experiments made by 
 
ii6 
 
 WATER SUPPLY. 
 
 Mr. J. C. Hoadley'* show that the slope is the same whether the 
 given yield of water be from an open well or from a driven tube ; 
 and that, with the same delivery, the level of the water in the 
 well or pipe is lowered to the same point. The particular exper- 
 iment alluded to was performed as follows : A 3-inch open pipe 
 was driven in pervious soil, into, and considerably below the sur- 
 face of the ground water, forming an open well — the earth being 
 removed from within. Into this pipe a suction pipe of i^-inch 
 diameter was dropped and wedged into position, so as not to 
 close the opening of the 3-inch pipe. Water 
 was pumped for a certain time, as much as 
 the well would supply. Subsequently, by 
 means of a cap, the suction tube was se- 
 cured in the 3-inch pipe and the opening of 
 the latter pipe hermetically sealed. The 
 suction pipe thus became, practically, a 
 driven well. Under these circumstances, 
 every other condition being unchanged, the 
 yield of water was approximately the same 
 as in the previous case, the slight difference 
 which existed being in favor of the open 
 well. 
 
 It should be noted that in what has been 
 said above, the yield of a driven well is com- 
 pared with that of a dug well of no great 
 ^'°- ^3- diameter. If the well be increased in di- 
 
 ameter to 30, 50, or 100 feet, so that the distance c c' (Fig. 22), 
 becomes considerable, the limit of measurable effect on the ground- 
 water level will be removed farther from a, and the yield of the 
 well will be appreciably greater — or, in other words, with the 
 same delivery, the point c will not fall so low. 
 
 With regard to the absolute amount of water which can be 
 utilized in a given locality, common sense, as well as science, tells 
 us that the amount of water which a given deposit can furnish 
 must be a definite quantity, although to us unknown ; and al- 
 though the driven wells may enable us to obtain this water more 
 conveniently than other methods, the absolute amount obtain- 
 able is no greater, and the supply cannot be inexhaustible, as 
 
 * Private communication : this statement is absolutely true only when the wells 
 are of the same diameter. 
 
NATURAL FILTRATION. 
 
 117 
 
 some of the enthusiastic advocates of the driven-well system 
 would have us believe. One great advantage which the driven 
 wells possess is the facility vi'ith which they may be sunk for 
 experiment or tempbrary use. For example, they were exten- 
 sively used by the British army in the Abyssinian expedition, 
 1867-68 ; and hence, in England, they are frequently called Abys- 
 sinian wells. 
 
 It should also be noted that the driven well ordinarily takes 
 its water from a lower point than that to which a dug well would 
 be sunk in the same locality. For this reason, a driven well may 
 continue to furnish water when a neighboring dug well has be- 
 come dry, and thus the impression that the driven well is inex- 
 haustible gains ground. 
 
 On this account also, the driven well is somewhat less liable 
 to pollution than a dug well in the same locality, as the polluting 
 material may be rather more diluted in the mass of the ground 
 water. Moreover, as has been said, the driven well often passes 
 through an impervious stratum of clay, so that the water ob-- 
 tained is entirely distinct from the ground water of the locality. 
 In this way good water may sometimes be obtained where the 
 surface conditions are very unfavorable, but there is always an 
 element of risk involved in sinking a well among sources of pol- 
 lution. 
 
 "Natural Filtration." 
 
 The ultimate source of the ground water is the rain, and that 
 the rain is the proximate source of the water obtained from a 
 well sunk into a gravel deposit far removed from any stream or 
 pond, scarcely any one can doubt. Such a well is that in Pros- 
 pect Park, Brooklyn, alluded to on page iii. This well is nearly 
 two miles from tide-water, and, although the natural level of the 
 water has been lowered by pumping, it is still a number of feet 
 above tide level. Generally, however, a gathering well, basin, or 
 gallery is located near a lake or river. This location is chosen 
 mainly because at such a place there is almost always a decided 
 movement of the ground water toward, or in the same direction 
 as, the stream ; but such a location is also chosen in order that 
 the river may make up any deficiency caused by the removal of 
 the ground water. 
 
 It was formerly supposed, and is so even now, by many per- 
 
Il8 WATER SUPPLY. 
 
 sons who have not made a study of the subject, that in such 
 cases the water is derived directly from the river, and filtered by 
 passing through the intervening sand and gravel. Undoubtedly, 
 in some cases, a considerable proportion is thus derived, but, as a 
 rule, the contrary is true, and, where the location is such that 
 most of the water must come from the visible body of water, 
 the supply generally proves inadequate. The beds of ordinary 
 streams furnish a poor filtering surface, and the experience with 
 artificial filters shows how soon an originally clean surface be- 
 comes clogged.* 
 
 That the view just expressed is correct, appears from a va- 
 riety of considerations. From the discussion of the effect of 
 pumping on the ground water (pages 109-112), it is evident that, 
 from a well situated near a stream, a certain amount of water 
 can be drawn without calling upon the stream at all for supply; 
 if, however, the circle of influence includes a portion of the 
 stream, some of the water may come from this source, unless, as 
 is indeed generally the case, it is easier for the water to come a 
 greater distance through open water-bearing deposits than to 
 force its way through the silted-up and more or less impervious 
 bed of the stream. 
 
 If we consider the character of the water, there are certain 
 general facts that are at once and readily noticeable : the water 
 thus obtained is generally clear and colorless ; it is of a quite 
 uniform temperature, cool therefore in summer, and in winter 
 much warmer than the water of neighboring ponds and rivers, 
 which, of course, approach in temperature very close to the 
 freezing point ; the water also differs in chemical character from 
 that of neighboring streams or ponds, generally being somewhat 
 harder. 
 
 With regard to the temperature, the difference is very marked, 
 even where the water is collected in an open basin and thus 
 exposed to the heating (or cooling) influences of the air. For 
 instance, in the filtering gallery at Lowell, Mass., during the 
 month of September, 1873, the highest temperature was 50° F., 
 the lowest 49° F. ; during the month of October, observations 
 
 * The term natural filtration is objectionable only so far as it implies that the 
 water is obtained from the lake or stream by a process of filtration : that the rain fall, 
 insj upon the ground may be said to be filtered naturally by passing through the inter- 
 slices of the water-bearing deposits is, of course, true. 
 
NATURAL FILTRATION. Ug 
 
 were made on thirteen different days showed identically the 
 same temperature, namely, 50° F. Between September 6 and 
 January i, the highest recorded observation is 52° F., on No- 
 vember 8, and the lowest is 47" F., December 31. There is no 
 corresponding record of the temperature of the river, nor is such 
 necessary, as every one knows that river water varies with 
 the temperature of the surrounding air, and in December must 
 have been nearly at the freezing point. At Waltham, Mass., 
 where the water is taken by means of an open shallow basin, 
 more marked differences have been observed between the tem- 
 peratures at different seasons. Thus in winter, when the river 
 was frozen, the temperature in the basin was about 44° F., and 
 the average of nineteen observations made at intervals from 
 August 23 to August 26, showed for the river an average tem- 
 perature of 74°. I F., and for the basin water an average tempera- 
 ture of 62°. S F. Such instances might be multiplied indefinitely, 
 and it seems quite impossible to account for the observed differ- 
 ences by the continuous passage of water through 100 feet or so 
 of gravel. In fact where no such differences are observed, it may 
 be a sign that the water does come from the stream, and the 
 water is likely to be otherwise unsatisfactory ; thus the city of 
 Toulouse, in, France, is supplied by a number of filtering gal- 
 leries in a gravel deposit on the banks of the Garonne. The 
 original gallery was built in 182- at a distance of about 60 meters 
 (200 feet) from the river. This furnished water acceptable in 
 quality, but deficient in quantity ; an increase of the length of 
 the gallery failed to furnish a corresponding increase in quantity 
 of water obtained. A second filtering gallery, or rather series of 
 connected wells, was constructed nearer to the river, at a distance, 
 in fact, of only ten meters. In this case, the water obtained man- 
 ifestly did come, in part at any rate, from the river : the water 
 was somewhat turbid, and what is very instructive, the passage 
 through a bank of thirty feet, and admixture, of course, with 
 some ground water, failed to bring the water to anything like 
 the uniform temperature of the other galleries. The tempera- 
 ture fell in winter to 2'' C. (35°.6 F.), and in summer rose above 
 21° C. (70° F.).* This gallery was therefore abandoned, and 
 others constructed at a greater distance from the stream. These 
 
 * D'Aubisson. Annales des Fonts et Chaussees, 1838. 
 
120 WATER SUPPLY. 
 
 furnish water which is satisfactory, except when in time of flood 
 the river covers the whole territory in which the galleries are 
 built, and the galleries become filtering galleries in the true 
 sense. 
 
 As marked differences as in the matter of temperature are 
 also observed in the chemical character of the water. This often 
 appeals to the eye by the absence of color in the (so-called) fil- 
 tered water, while the water of the rjver may be strongly colored ; 
 or, if the gallery be alongside of a pond, the latter may be filled 
 with algse in a state of decomposition, without producing the 
 slightest effect on the gallery water. It is, however, the hardness 
 of the water which generally attracts attention, being noticed 
 where the water is used for washing or in steam boilers. Usu- 
 ally the ground water is harder than the surface water of the 
 same region, but occasionally the reverse is true. Belgrand gives 
 a number of examples from French localities, from which may 
 be cited the following : * 
 
 Water of Rhone, at Lyons 16° 
 
 Water of filtering gallery at Lyons 17-94 
 
 Water of Loire, at Nevers 4.96 
 
 Water of collecting well 20. 70 
 
 Water of Loire, at Blois 7. 76 
 
 Water of the gallery (which is beneath the bed of the river) 14.45 
 
 Sharpies has found f that the water in the filter gallery near 
 Little Pond, Cambridge, contains nearly twice as much lime as 
 that of the pond, and instances might be multiplied indefinitely. 
 In the case of the Dresden water supply the river water is harder 
 than that obtained from the collecting wells. % 
 
 Even when the gallery or well is sunk directly in the bed of 
 the river, or in an island surrounded on all sides by the river or 
 pond, the ground water still contributes largely or wholly to the 
 supply. Many experiments have shown that the water in a 
 gravel deposit directly beneath a river differs essentially from 
 that of the stream itself. 
 
 The behef that a well or gallery located near a pond or stream 
 
 * La Seine, etc. , pp. 463 and following. 
 
 f Twelfth Annual Report of the Cambridge Water Board, for the year 1876. 
 Boston, 1877 ; page 30. 
 
 % Salbach. Das Wasserwerk der Stadt Dresden ; 3r Theil, page 7. 
 
NATURAL FILTRATION. 121 
 
 does not necessarily derive its supply from the visible body of 
 fresh water, finds confirmation in tjie well-known fact that springs 
 are often observed to issue from the sand along the sea shore, 
 even below low-water mark,* and fresh water is often obtained 
 by sinking wells very near the shore. Generally, in such cases, 
 the surface of the ground and the water table rise as they recede 
 from the shore, the ground water, derived from the rain, passing 
 with more or less resistance to the sea. In such wells, the 
 water rises and falls with the tide, as the water must enter the 
 sea under the pressure of a varying height of salt water, but the 
 salt water itself does not penetrate the soil and reach the well 
 itself. From such wells a certain amount of water can be. 
 pumped. If the amount pumped exceeds that which the ground 
 water can furnish, salt water may then' be drawn into the con- 
 tributing area, and the water become brackish. Even where the 
 ground near the sea is level, the mere effect of the rain falling 
 upon the sandy area is sufficient to create a deposit of fresh 
 water which may crowd out or prevent the entrance of salt water. 
 Darwin, in the voyage of the " Beagle," discovered this to be 
 the case in low coral islands of the Pacific, close to the sea ; and 
 in Holland, Amsterdam, the Hague and Leyden obtain their 
 water from collecting canals in the sand-dunes which form an 
 almost barren strip of country from 2 to 5 kilometers wide, hav- 
 ing only a few elevations of surface. Sometimes fresh water 
 overlies the salt, so that shallow wells furnish fresh water, while 
 deeper wells give brackish or salt water. McAlpine has made 
 the interesting observation that where, as on Long Island, N. Y., • 
 the water table slopes down to the sea, the underlying deposit 
 of salt water slopes away from the sea — the higher and conse- 
 quently heavier column of fresh water at some distance inland 
 being able to displace the salt water to a greater depth : thus, 
 the vertical section of the body of fresh water, in the direction 
 of its flow, would be that of an elongated wedge. Salt water 
 has been found underlying the fresh water in other localities — 
 
 * A remarkable example of this occurs at the four iron forts at Spithead, Eng. 
 Here wells are sunk on artificial islands, at a considerable distance from the shore, 
 and, although two of them are over 550 feet deep, they pass entirely through sand and 
 gravel. In spite of this location, the water of the wells contains only a small propor- 
 tion of chlorine (18.6, 11. 4, 4 i, 7.6 parts in 100,000 respectively!, showing that 
 almost no sea water finds its way into the wells. — The Analyst, April, 1883. 
 
122 WATER SUPPLY. 
 
 for instance, at Hull, in England * — where the salt water is sup- 
 posed to be due to the infiltration of sea water, and not to the 
 mineral character of the rock. 
 
 Preliminary Examination of a Proposed Ground Water Supply. 
 
 The least satisfactory point in connection with ground-water 
 supplies is that the amount of water to be obtained in any one 
 locality is limited in amount, and it is very difficult to tell in 
 advance how large an amount a given region can be relied 
 upon to supply, except as a result of thorough surveys and long- 
 continued experiments. 
 
 The effect of pumping upon the level of the ground water 
 and information as to the direction of its flow may be obtained 
 by driving a number of iron pipes with perforated " points " at 
 regular distances, preferably in two lines at right angles to each 
 other, intersecting in the experimental well. Observations should 
 be made on the natural level of the ground water before pump- 
 ing is begun ; and the pumping is best conducted by keeping the 
 level of the water in the well at a constant distance below the 
 natural level of the ground water, or below the level of the water 
 in the pond or stream. Although absolute equilibrium cannot 
 be established for a considerable time, unless thfe water comes 
 very freely, and in the absence of rain, sufficient indications can 
 be obtained to form judgment, within limits, as to the probable 
 yield of the well. In locating a gallery or elongated basin, refer-, 
 ence will be had to the direction of the greatest movement of 
 the ground water, which is sometimes in the direction of the flow 
 of the visible stream and sometimes at right angles to it. 
 
 A preliminary examination with reference to a future supply 
 should include a careful survey of the entire drainage area, and 
 in all cases the preliminary examinations should be made by 
 those conversant with the matter, as there is great liability to 
 overestimate the probable yield of water. As a rule, the amount 
 obtained from any such well is greater at first, as it requires time 
 to drain out the water naturally occupying the territory which 
 hereafter is to flow into the well. On the other hand, the effect 
 of the draught of water toward a single point is to open chan- 
 nels in the porous material so that in some cases the yield 
 
 * Proc. Inst. Civ. Eng. Gr. Br., Iv, p. 257. 
 
EXAMINATION OF GROUND WATER. 1 23 
 
 increases with time. Besides an assurance that the quantity of 
 water obtainable is and will be sufficient, it is necessary to know 
 that the water is satisfactory in quality. As far as the character 
 of the water is concerned, it is in New England generally good 
 when sufficiently abundant ; it is almost always harder than the 
 river water, but in most localities this difference in hardness is 
 small, although appreciable. In limestone regions, however, the 
 ground water is often so hard as to be unsuited for use ; and 
 sometimes the presence of streaks or beds of clay or of ochre 
 makes it impossible to obtain clear water. 
 
 The absence of such injurious deposits should be ascertained, 
 not only at the point at which it is proposed to locate the actual 
 well or gallery, but also in the immediate neighborhood, espe- 
 cially in the direction in which the gallery is likely to be ex- 
 tended or where additional wells may be sunk. 
 
 Leipzig, in Germany, has had a ground-water supply since 
 1866. The water, which was of good quality, proved insufficient 
 for the wants of the city, and the supply was increased (1871-72) 
 by the construction of an additional collecting gallery. Appar- 
 ently the work was done without sufficient preliminary examina- 
 tion, for the works were scarcely opened before trouble was 
 experienced, and it was found that the locality into which the 
 gallery had been extended was generally unsuitable. There was, 
 however, one peculiar and instructive difficulty. The gravel of 
 the deposit in which the gallery was located contained — as grav- 
 els frequently do — -oxide of iron. This ordinarily would give no 
 trouble, but the gallery intersected an old river-bed containing 
 many partially decayed stumps and other organic matter which, 
 in the presence of water, reduced the oxide of iron to the pro- 
 toxide condition, forming soluble protosalts of iron. These com- 
 pounds dissolving in the ground water, found their way into the 
 collecting gallery in large quantity. As soon as these soluble 
 protosalts of iron come into contact with the air they are oxi- 
 dized, and a deposit of the red hydrated oxide is formed. In 
 Leipzig the oxidation generally took place before the water 
 reached the consumers, and the complaint was with reference to 
 the muddy, red appearance of the water when drawn ; this was 
 easily overcome by filtration. Sometimes, however, the water 
 reached the consumers before the oxidation was complete, in 
 which case the filtered water tasted " like ink," and on standing 
 
124 
 
 WATER SUPPLY. 
 
 deposited a further quantity of a red sediment.* It may be 
 stated that the remedy in this case consisted in seeking a new 
 supply in a more favorable locality. 
 
 As we have seen, even where a well or gallery is located near 
 a stream or pond, the proportion of water received from the 
 visible stream or pond is usually small ; therefore, to • obtain in- 
 formation as to the character of the water to be obtained, it is 
 much more important to examine the ground water than the 
 water of the river. The examination of the latter should not, 
 however, be neglected, and it would scarcely ever, if ever, be 
 advisable to locate " natural filtration " works on the banks of a 
 stream which was seriously polluted. 
 
 Further, although there is less liability to pollution than in 
 the case of small shallow wells sunk near dwellings, slaughter 
 houses, factories, or stables, it must be remembered that the 
 ground water is fed by the percolation into it of the atmospheric 
 water, and that it is possible to pollute even a large body of water. 
 This fact should be taken into account in choosing a locality for 
 the collecting wells. 
 
 TABLE XVII. — Examination of Ground Water. 
 [Results expressed in Parts per 100,000.] 
 
 Locality. 
 
 Ayer, Mass, 1880 
 
 Newton, Mass., basin near Charles River, 1877, 
 
 Taunton, Mass., basin, Aug., 1877 
 
 " " river, '^ *' 
 
 Waltham, Mass., basin, Dec, 1873 ■ • 
 
 " " river, " '^ 
 
 Lowell, Mass., gallery, Jan., 1874 
 
 '' '* river, " " 
 
 Cambridge, Mass., gallery, Dec, 1876 
 
 '' " pond, '* '' 
 
 Chautauqua, N. Y., filter chamber 
 
 Lake 
 
 Indianapolis, Ind., well, 1880 
 
 " White River, " 
 
 3-3 
 3-9 
 
 5-6 
 S.8 
 
 6.5 
 
 5-7 
 
 6.4 
 4-S 
 
 18.6 
 14. o 
 
 10.2 
 7.0 
 
 34.0 
 2g.o 
 
 S 
 
 s 
 
 0.009 
 0.005 
 
 0.005 
 0.006 
 
 0.006 
 0.005 
 
 0.080 
 0.070 
 
 0.003 
 0.005 
 
 CQ S 
 
 0.008 
 0.002 
 
 O.OIO 
 
 0.021 
 
 0.006 
 0.016 
 
 0.003 
 0.010 
 
 0.005 
 0.016 
 
 0.005 
 0.006 
 
 Authority. 
 
 W. R. Nichols 
 J. M. Merrick 
 
 W. R. Nichols 
 
 S. P. Sharpies 
 S. A. Lattimorc 
 05'T. C. Van Nuys 
 
 * Hofmann, Dr. Franz : Die Wasserversorgung zu Leipzig. Pph. 8vo, pp. 62. 
 Leipzig, 1877. 
 
EXAMINATION OF GROUND WATER. 12$ 
 
 In Table XVII are given the results of the partial analyses 
 of some ground-water supplies, together, in most cases, with the 
 analysis of the neighboring pond or stream. The chemical ex- 
 amination of a water under discussion directs itself, mainly, to 
 proving the freedom from organic matter and other signs of pol- 
 lution, and to ascertaining that the hardness is not excessive. 
 
 In this connection we may mention a peculiar trouble which 
 has occurred at several foreign water works.* 
 
 Since September, 1877, ^ portion of the Berlin water supply 
 has been taken from the neighborhood of the "Tegeler See," by 
 means of a series or line of 23 wells running parallel with the 
 shore of the lake. 
 
 Shortly after the introduction of the water, complaints arose 
 as to its quality, and investigation proved the difficulty to be 
 twofold. It is frequently noticed that water — and especially 
 water from a driven well — although apparently clear when first 
 drawn, becomes turbid on standing and deposits an ochreous sedi- 
 ment. This is generally due to the presence in solution of the 
 protocarbonate or to some organic protosalt of iron, which — on 
 exposure to the air — becomes oxidized and changed to an insoluble 
 hydrated sesquioxide. This was the cause of the trouble which 
 occurred at Leipzig, and this was one of the difficulties with the 
 Tegel ground water, but the microscope showed that the ochre- 
 ous sediment which settled from samples of the water, and which 
 accumulated in the reservoirs and in the pipes, especially in " dead 
 ends," was by no means made up wholly of amorphous mineral 
 matter, but consisted very largely of algcs, dead and alive. 
 
 Most noticeable among the algm was the Crenothrix Kuhni- 
 ana {Crenothrix polyspora, Cohn). This plant was first discovered 
 by Kiihn in 1852, in the drains of a cultivated field in Silesia, but 
 has since been found in wells in various parts of Europe, and is 
 probably very widely distributed. 
 
 In Berlin, it was found in the wells, in the reservoirs and in 
 the service pipes, in various stages of development and decay. 
 The spores are minute spherical or oblong bodies from one one- 
 thousandth to six one-thousandths of a millimeter in diameter. 
 From these spores, and by other means of development, the 
 
 * This is abridged from an account of the trouble given by the author in the Jour- 
 nal of the Franklin Institute, March, 1882. 
 
126 WATER SUPPLY. 
 
 plants grow into coniparatively long threads, each of which on 
 examination is seen to be made up of a number of individual 
 cells, end to end, inclosed eventually in a gelatinous sheath. The 
 general appearance of a mass of these threads is shown in the 
 figure, and the masses are sometimes a centimeter or more in di- 
 ameter. 
 
 Fig. 24. — Crenothrix Kuhniana. 450 : i. 
 
 The threads are at first, like the spores, transparent and 
 colorless, but by the absorption of iron in some form or other 
 they become colored from olive-green to a dark brown. They 
 eventually, in many cases, become incrusted with the hydrate of 
 iron to such an extent that their structure becomes invisible, 
 but it may be made evident by dissolving away the hydrate of 
 iron by very dilute chlorhydric acid. Under favorable circum- 
 stances the plants may develop with great rapidity, and Pro- 
 fessor Kiihn speaks of their having frequently stopped up agricul- 
 tural drain pipes. Also, the pipes in which water is taken from 
 a well ten meters deep, in the neighborhood of the Plotzensee, 
 near Berlin, have in summer been choked and nearly filled up by 
 the multiplication of the same organisms. In the reservoirs and 
 in the " dead ends " of the service pipes they seemed to accumu- 
 
EXAMINATION OF GROUND WATER. 
 
 127 
 
 late by growth as well as by deposition. While the plants 
 develop more rapidly in the warm season, they are found at all 
 times of the year in all stages of development. 
 
 It may be remarked, in this connection, that the Crenothrix 
 has great vitality ; thus, Dr. Zopf exposed a quantity in water 
 out-of-doors from the first of January to the middle of February. 
 The water was, of course, frozen, and during the time the tem- 
 perature fell as low as to — 8°R. (ii°F.), but after being thawed 
 out the plants had, in a few weeks, contrary to all anticipation, 
 revived again or new ones had grown from the spores. 
 
 The Crenothrix seems to live and develop in the ground itself, 
 and in an examination which was made of the water from a 
 number of wells in different parts of Berlin, the same plant was 
 found in many cases, in one instance at a depth of more than 24 
 meters from the surface. 
 
 Whether its presence would be revealed in the preliminary 
 examination of a ground water is doubtful, but it ought certainly 
 to show itself if pumping experiments were carried on for any 
 considerable length of time. 
 
 There seems to be no remedy for this trouble. It was found 
 possible in Berlin to filter the water artificially through sand — 
 ' after exposing it to the air — so as to obtain the supply perfectly 
 clear ; but, of course, the filters were very much fouled, and, on 
 account of the difficulty of washing the sand thoroughly and the 
 risk that the spores of the plant would eventually find their way 
 into the lower part of the filters and thence into the service, it 
 was thought best by those in charge of the works to abandon 
 the wells altogether, and to make use of water taken directly 
 from the lake and filtered in the usual manner. The same 
 trouble occurred in Halle, and it is stated that it was overcome 
 by sinking other wells in a different locality. In the second 
 locality the water was much harder and free from the Crenothrix ; 
 in fact, when it was mixed with water from the previous source 
 it brought about the extermination of the plant ; hence it has 
 been inferred that the presence of a considerable amount of car- 
 bonate of lime is fatal to the plant, but this is very doubtful in 
 view of what follows. 
 
 The same trouble has occurred recently at Lille, in France. 
 The source of supply is here a subterranean reservoir in marl and 
 water-bearing chalk lying near the surface. The water comes to 
 
128 WATER SUPPLY. 
 
 the surface in actual springs which originate at no great depth. 
 The hardness of the water is about 25' (French), and the ex- 
 amination made in 1864 showed 44 parts of total solids in 
 100,000, a large proportion being carbonates of lime and mag- 
 nesia. The water was then considered of good quality,* but for 
 some time there has been complaint of a red color and of an un- 
 pleasant taste and odor. The matter becoming very serious in 
 the spring of 1882, led to the discovery that the trouble was 
 mainly due to the Crenothrix. The previous winter had been 
 very dry, and the water level had been lowered about five meters. 
 The rains of the spring raised the water level, and seem to have 
 washed out the plants into the sources of supply ; f it is possible 
 also that contamination of the overlying soil had increased the 
 amount of soluble iron salts which are necessary to the growth 
 of the Crenothrix. 
 
 The Pollution of Domestic Wells. 
 
 In isolated dwellings and in villages and small towns not yet 
 provided with a public water supply, drinking water must, as a 
 rule, be obtained either by collecting the rain water and storing 
 it in tanks and cisterns, or else by sinking wells. On account of • 
 the clearness and nearly uniform temperature of the ground wa- 
 ter, the latter method is usually preferred when practicable. In 
 the majority of pases the location of the well is dictated simply 
 by convenience, and it frequently happens that it is in close prox- 
 imity to a privy, or to cesspools, or to a barn or stable. The result 
 is that the well is very liable to pollution, and, more often than 
 not, it is simply a question of time when the water shall become 
 unfit for use. The pollution of the well generally takes place 
 gradually. The ground gradually becomes charged with the 
 soakage from the privies and manure heaps, and percolating rain 
 water carries the impure matter into the ground water from 
 which the well draws its supply. In other cases, actual channels 
 are formed, by which the foul liquid trickles or flows into the 
 well itself,, or a leaky drain, laid near the well, may be the source 
 of the trouble. 
 
 Whatever views may be held of the effect upon the human 
 
 * Masquelez : Ville de Lille, ifitablissement de la Distribution d'Eau, Paris, 1879. 
 f Alf. Giard : Comptes rendus, xcv (1882), 247-249. 
 
POLLUTION OF WELLS. 
 
 129 
 
 system of drinking such water, there is no question whatever as 
 to the pollution itself, and although the water may appear clear 
 and bright, and be inoffensive to the senses, chemical examina- 
 tion may show that it is highly charged with the products of de- 
 composition. Moreover, there are hundreds of cases on record 
 where sickness h^s been coincident with the use of polluted well 
 water, and, although the evidence is of necessity circumstantial 
 (see Chapter I), it is too striking to be disregarded. In the 
 present state of knowledge, it must be said that the continued 
 use of a well water proved to be polluted is as unjustifiable as 
 suicide generally is. 
 
 It is often difficult to persuade the owner of a polluted well 
 to abandon its use. The water tastes good and has been used 
 for years without producing any bad effects. Meanwhile, how- 
 ever, in these years the neighborhood has become thickly settled, 
 the various possible sources of contamination have increased, 
 and the whole ground water of the region has felt the effect. At 
 the same time, in urging the abandonment of the well, one cannot 
 say that, in spite of the pollution, it may not be used for years 
 more without noticeable ill effects. Under what conditions the 
 water may become injurious, and when, no one can say. 
 
 It is also difficult to realize the distance from which the pol- 
 lution may come. Until the water of the well becomes contam- 
 inated to a very great extent, the taste gives no evidence of con- 
 tamination, but occasionally accidental evidence is furnished of 
 the distance from which communication with the well may 
 exist. 
 
 An illustration of this point, and ■ a further illustration of 
 certain chemical changes which have been already alluded to 
 is the following. In Wernigerode, Germany,* a certain well 
 which had always been nearly free from iron, suddenly began to 
 furnish a chalybeate water. Clear when drawn, the water soon 
 became turbid, and deposited on standing a copious ochrey sedi- 
 ment. It was finally discovered that this sudden change was due 
 to the emptying of several casks of spoiled beer into the ground 
 at a distance of some 35 meters (115 feet). The organic matter 
 thus introduced into the ground acted as a reducing agent on 
 the ferric oxide contained in the soil, and the iron, dissolved as 
 
 * Wockowitz, E. : Wemigerode's Trinkwasser. Wernigerode, 1873. 
 9 
 
I30 WATER SUPPLY. 
 
 protocarbonate, found its way into the body of water from which 
 the well was supplied. 
 
 In some places, owing to the very nature of the locality, 
 shallow wells are to be rejected as sources of supply. Thus, 
 Dr. Smart * says, with reference to New Orleans : " The well 
 waters of New Orleans are unfit for use. They are but Uttle 
 less impure than the sewage water carried off by the drainage 
 canals, yet they are reported as being employed for family use, 
 in bakeries, and for stock, especially in summer, when the cistern 
 supply fails. The site of the city is waterlogged to within a 
 few feet of the surface. One well, on Chestnut street, the least 
 impure of those examined, is only lo feet deep, and contains 7 
 feet of water. The saturated soil is of great depth, and the 
 ground water is practically stagnant. The filtration into the 
 wells is insufificient even to free the water from turbidity. Or- 
 ganic matter is unaffected by the process. The water contains 
 alkaline carbonates, chlorides, large amounts of free ammonia, 
 but no nitrates or even nitrites. In four wells examined, the 
 ammonia from organic matter amounted to 0.039, O.041, 0.044, 
 0.080 part ; while in the sewage from the Orleans canal it only 
 reached 0.120 part. These samples are so impure that the use 
 of well water in New Orleans should be interdicted.! Even care- 
 , ful filtration should not be relied on to purify such waters. Fil- 
 tration is not a process by which dangerous waters may be util- 
 ized, but simply a guard against the possibility of danger in 
 doubtful waters." 
 
 We have thus far spoken only of wells which are sunk into 
 the ground water. These are the most common, but many wells 
 are sunk into a more or less compact rock, and the water comes 
 through fissures in the rock. In such cases it is often difficult to 
 tell where the water does come from, and the well is liable to 
 contamination from distant sources. The pollution is liable to 
 be even more serious than in wells sunk into the ground water, 
 because the contaminating substances carried by the stream of 
 water do not have the same opportunity to be oxidized as they 
 do when the water passes with comparative slowness through a 
 body of sand or gravel. 
 
 * Bulletin National Board of Health, April 17, 1880. 
 t Such use is now prohibited by law, J883. 
 
EXAMINATION OF WELL WATERS. I3I 
 
 Of the well waters which are submitted to chemical examina- 
 tion a limited number show by the absence of ammonia, nitrog- 
 eneous organic matter, and chlorides in appreciable quantity, 
 that they are free from all contamination ; on the other hand, a 
 considerable proportion (not, however, one-half according to the 
 experience of the author) may be condemned at once ; the re- 
 mainder can only be considered doubtful or suspicious. In the 
 cases of those suspicious wells which cannot be absolutely con- 
 demned, the proper course is to have, for a time at least, some- 
 what frequent examinations made of the water to see whether 
 the impurity is on the increase. For this particular purpose, it 
 is usually sufBcient to follow a single ingredient, say, for exam- 
 ple, the chlorine existing as chlorides. If the amount of chlorine 
 increases to any considerable extent, the source of impurity 
 should be ascertained and the water be protected therefrom, if 
 possible, or else be rejected from use. 
 
 It may be said, in a general way, that a good well water 
 should not contain over 0.005 P^^ of ammonia in 100,000, or 
 over o.oio part " albuminoid ammonia," and, in most places, not 
 over i.o part of chlorine (as chlorides). The amount of solid 
 matter in solution depends necessarily upon the locality, and 
 what might be a reasonable amount in one region would be very 
 abnormal in another. The presence of nitrates is also suspicious, 
 but, unless the quantity is very considerable, cannot alone con- 
 demn the water. Dr. Charles Smart, U. S. A., accepts o.oio 
 albuminoid ammonia in 100,000 as suspicious, and 0.015 as a limit 
 in the case of well waters " in the denser settlements, and in every 
 case where an animal origin to the organic matter is indicated by 
 careful survey or chemical analysis." 
 
 In the case of doubtful waters, the greatest satisfaction may 
 be obtained when it is possible to find in the same immediate 
 neighborhood a well of whose freedom from contamination there 
 can be no doubt. The comparison of the waters of the two 
 wells will probably enable one to decide the question as to the 
 contamination of the first well. The next most satisfactory 
 course of procedure is to throw a quantity of salt (or brine) into 
 the various cesspools, drains, etc., and to determine at frequent 
 intervals the amount of chlorine (as chlorides) in the water of 
 the well. As instances in which this method was used with good 
 results, the two following may suffice : 
 
132 
 
 WATER SUPPI^y. 
 
 In No. I, the well was located lOO feet from the privy ; a 
 bushel of coarse salt was put into the privy-vault October 24, and 
 a bushel of fine salt on October 31. This caused an evident 
 increase of the amount of chlorides in the well, as appears from 
 the figures. In case No. II, two bushels of salt were put into a 
 cesspool which was 75 feet from the well. A sample of water 
 was then taken and afterward at intervals of three days. The 
 effect on the well water was not as marked as in No. I, but 
 the results were confirmed by a subsequent examination of the 
 locality : 
 
 Date of Examination. 
 
 as Parts in 
 
 October 17 
 
 3-3 
 
 26 
 
 3-9 
 
 29 
 
 3-9 
 
 November 2 
 
 4.0 
 
 5 
 
 4-4 
 
 8 
 
 3-5 
 
 13 
 
 3-4 
 
 17 
 
 3-4 
 
 20 
 
 3-3 
 
 23 
 
 3-1 
 
 II. 
 
 II. 
 
 No. of Sample. 
 
 Chlorine 
 
 I 
 
 1.4 
 
 5 
 
 1.5 
 
 3 
 
 1.6 
 
 4 
 
 i-7 
 
 5 
 
 1-7 
 
 6 
 
 1-9. 
 
 7 
 
 2.6 
 
 8 
 
 2.0 
 
 9 
 
 2.0 
 
 10 
 
 1.9 
 
 In Table XXIII are brought together the results of the 
 examination of a few well waters from various' localities. The 
 table might be extended to an indefinite length, as the reports 
 of boards of health and water committees would, in almost 
 every case, contribute to the list. 
 
 TABLE XVIII.—ExAMiNATioN OF Well Water. 
 
 [Results expressed in Parts 19 100,000.] 
 
 Locality. 
 
 Sauffus, Mass 
 
 *' Another 
 
 Williamstown, Mass., 
 
 '* Another 
 
 " Another 
 
 North Adams, Mass.. 
 
 Gloucester, Mass 
 
 Watertown, N. Y 
 
 Croton Falls, N. Y.... 
 
 Lockport, N. Y 
 
 Southampton, L. I.. . 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 4 
 
 5 < 
 
 u 
 
 
 p 
 
 i 
 
 E 
 
 
 z 
 
 X 
 
 Quality. 
 
 h 
 
 < 
 
 u 
 
 
 7 3 
 
 O.OOI 
 
 0.002 
 
 I.I 
 
 Good. 
 
 21.0 
 
 0.002 
 
 0.002 
 
 3 1 
 
 Suspicious 
 
 19-3 
 
 0.002 
 
 o.oog 
 
 1-7 
 
 " 
 
 63.1 
 
 0.006 
 
 0.015 
 
 10.2 
 
 Polluted. 
 
 112. 1 
 
 0.005 
 
 0.013 
 
 40.0 
 
 " 
 
 is:? 
 
 0.014 
 
 0.009 
 
 2.2 
 
 *^ 
 
 0.230 
 
 0,029 
 
 9-S 
 
 '* 
 
 37-4 
 
 0.002 
 
 0.005 
 
 9-3 
 
 Good. 
 
 13.2 
 
 
 
 
 
 0.8 
 
 '■ 
 
 96.8 
 
 O.OOI 
 
 0.007 
 
 8.9 
 
 Doi'btful. 
 
 45 -o 
 
 o.oq6 
 
 0.018 
 
 4.0 
 
 Bad. 
 
 Authority. 
 
 W. R. Nichols. 
 
 E. Waller. 
 
CHAPTER VII. 
 
 DEEP SEATED WATER AS A SOURCE OF SUPPLY. 
 
 While a portion of the rainfall which soaks into the ground 
 soon encounters an impervious stratum, above which it collects 
 to form the ground water of the locality, much of the water pre- 
 cipitated from the atmosphere falls upon the edges of upturned 
 rocky strata, or upon rock deposits which are either themselves 
 porous pr so fissured that they afford a more or less free passage 
 for water. When the pervious stratum has an outcrop at some 
 lower level, the water may issue in the form of springs, more or 
 less copious. Where the course of the water has not been too long, 
 and it has not, consequently, taken up a large amount of mineral 
 matter, such springs furnish one of the best sources of drinking 
 water, althoiigh the water is very often, in fact usually, less well- 
 suited for technical purposes, on account of its hardness. The 
 advantage of spring water over surface water for drinking is con- 
 sidered by some so great as to justify the incurring of very con- 
 siderable expense in order to procure it. Thus, the city of Vienna 
 constructed extensive water works for the sake of bringing water 
 from springs which are sixty miles distant. 
 
 Artesian Wells. 
 
 When the water precipitated from the atmosphere is absorbed 
 by a pervious stratum which is situated between two impervious 
 strata, the water may exist under considerable hydrostatic press- 
 ure. The occurrence of a " fault " in the strata may allow the 
 water to rise to the surface of the ground as springs, but offen 
 the water can be utilized only by sinking or boring artesian wells. 
 
 An artesian well is a well which is sunk or bored through an 
 impervious stratum so as to reach a water-bearing stratum in 
 which the water is under hydrostatic pressure ; so that, as soon 
 as the well is opened, it rises through the impervious stratum 
 and often to, or higher than, the surface of the ground. Arte- 
 
134 
 
 WATER SUPPLY. 
 
 sian wells may, therefore, be regarded as artificially opened springs. 
 The term artesian is frequently applied to non-flowing deep wells, 
 but, while the question of flowing or non-flowing may be unes- 
 sential, the term is improperly applied to wells, however deep, 
 
 when the water is taken from the 
 deposit into which the well is 
 bored or sunk, and where the 
 water collects from the fissures 
 and cavities of the rock itself. 
 Thus, in England, there are many 
 deep wells in the chalk or in the 
 new red sandstone which collect 
 
 Fig. 25. — ARTESIAN WELL. j .M- i r i-t, t- 11 
 
 and utilize water from the chalk 
 or from the sandstone itself, and which are not properly to' be char- 
 acterized as artesian. In this country, driven wells are often called 
 artesian wells, and they may be properly so designated if they pass 
 through a stratum of clay, so that the water rises from an underly- 
 ing deposit not in communication with the ground or surface water 
 of the same locality : driven wells, however, as we have already 
 seen, often utilize simply the ground water where they are driven. 
 
 Although, within modern times, improvements have been 
 made in the methods and apparatus employed for boring wells, 
 wells of this description are of great antiquity. They are found 
 in China, and many such, wells have existed for a long time in 
 North Africa, in the oases of the Sahara. Here, until recently, 
 they were excavated by hand, the earth and other material being 
 drawn up in baskets. Finally, a thin stratum of rock was reached 
 beneath which experience had shown that water existed. This 
 rock stratum was cautiously perforated, and as soon as it was 
 pierced the workman was drawn up rapidly — and not always 
 safely — as he was sometimes overtaken by the rush of water. 
 The French in the province of Constantine (Algeria), between 
 the years 1856 and 1878, bored over 400 wells. The flowing wells 
 numbered 158, with an average depth of 85.5 meters; the tem- 
 perature of the water was generally betwen 21° C. and 26° C, and 
 the total solids between 300 and 600 parts in 100,000.* 
 
 There are a great many artesian wells in various parts of the 
 
 * Les Forages artesiens de la province de Constantine (Algerie). Resume des 
 Travaux executes de 1856 i 1878. Par M. Jus. Svo, pp. 97. Paris, Imprimerie 
 Kationale, 1878. 
 
ARTESIAN WELLS. 
 
 135 
 
 United States. Thus, Professor Winchell, in 1856,* mentions as 
 many as 74 such wells in a single and somewhat circumscribed 
 region of middle Alabama, and, of late years, many wells have 
 been sunk in the southwestern part of the country with favora- 
 ble results. 
 
 The sinking of artesian wells is attended with great uncer- 
 tainty as regards both the quality and the quantity of the water 
 to be obtained, and many wells have been sunk which have failed 
 to reach water at all, or from which only water unfit for any do- 
 mestic use has been obtainable. To judge of the probability of 
 success in sinking a well in a new locality, a knowledge of the 
 geological character of the underlying strata is essential. At 
 Charleston, S. C, where are several successful artesian wells,-)- the 
 possibihty of obtaining such wells was inferred from a knowledge 
 of the geology of the region. More than ICX) miles fronri the city, 
 starting from Augusta, Ga., and proceeding northeastwardly, a 
 granite ridge rises to the surface of the earth, exposed to view in 
 favorable positions, elsewhere covered with superficial drift sands 
 and clays. The line may be followed northward through North 
 CaroHna, Virginia, and Maryland. On the broad surface of this 
 granite ridge, and on its seaward slope, the sands drink in the 
 rain water that falls. The streams from the up country that 
 cross the ridge may also supply their quota. The water thus 
 imbibed sinks down by the force of gravity, ever seeking the low- 
 est attainable position. Now, the tertiary beds of the Charles- 
 ton Basin, the cretaceous beds under them, and any other sedi- 
 mentary beds beneath the cretaceous, must rest against this 
 eastern slope of the granite ridge, and their sandy layers must 
 drink in the water filtering through the sands. As all of these beds 
 have a gentle slope toward the coast, the water will follow them 
 down in their course. These formations, no doubt, continue 
 their course under water for many miles, and, indeed, there is 
 evidence that the water contained in them finds a discharge into 
 the sea. To this cause are attributed the springs of fresh water 
 that have been observed to rise, bubbling up at times in notable 
 quantity through the salt water at points along the coast, fifteen or 
 twenty miles from the shore. Moreover, in all the deep wells in 
 
 * Proc. Amer. Assoc, x (1857), p. 83. 
 
 t Municipal Report of the City of Charleston, S. C. Artesian Wells, 1881. 
 
136 WATER SUPPLY. 
 
 Charleston, varying from 60 to 1,260 feet in depth, the level or 
 head of water in the pipes has been observed to oscillate at tidal 
 intervals to an extent varying from 4 to 6 inches. The explana- 
 tion is simple. In issuing from its natural vent under the sea 
 the fresh water must lift the column of salt water above, the 
 weight of which acts as an obstruction. When the tide at sea 
 is high, this column is greater than it was at low tide. The con- 
 sequence is a diminution of the escape of water by that channel, 
 and a compensating increase of the discharge through other 
 channels not so obstructed, and an increase in the head of water 
 in the wells.* 
 
 One disadvantage of sinking artesian wells for town supply 
 is the great uncertainty as to the quality of the water, and 
 the fact that water from considerable depths is often of ele- 
 vated temperature, and therefore not fit to drink unless cooled. 
 Moreover, the water is apt to be charged with a large amount 
 of mineral matter derived from the strata through which it has 
 flowed or percolated, or in contact with which it has remained 
 for a long period of time. The widely-known well at Cre- 
 nelle, Paris, which is about 1,800 feet deep, has a temperature of 
 27° C. (8o°.6 F.), and contains only about 14 parts in 100,000 of 
 dissolved solids, whereas a well in St. Louis, Missouri, sunk at 
 the sugar refinery of Belcher and Brothers to a depth of over 
 2,000 feet, and at an expense of $10,000, furnishes about 75 gal- 
 lons per minute of water emitting a strong odor of sulphuretted 
 hydrogen, and containing 879.1 parts of dissolved matter in 
 100,000 parts ; this water is entirely useless for the purposes of 
 the refinery or for domestic use. As already stated, the artesian 
 wells in Algeria contain from 300 to 600 parts of dissolved 
 solids in 100,000 parts of water, and would, in most localities, be 
 at once rejected even for purposes of irrigation. In the absence, 
 however, of better water, such wells as these, are regarded as 
 godsends by the inhabitants. 
 
 The fact of a considerable amount of dissolved solids does 
 not necessarily prove that an artesian water is unfit for use, 
 although usually the salts present are objectionable in character. 
 Sometimes the dissolved matter, in the absence of lime, magne- 
 
 * The above statement is condensed from the Charleston Municipal Report 
 already cited. 
 
ARTESIAN WELLS. 1 37 
 
 sia and sulphates, may be unobjectionable in character, although 
 the large amount present may be undesirable. For example, the 
 water of the Wentworth Street artesian well in Charleston, S. C, 
 which contains 273.6$ parts of total solids in 100,000, has been 
 used for years. In this case the dissolved matter is almost 
 entirely common salt and carbonate of soda, and the use of the 
 water is held to be beneficial in dyspepsia and kindred diseases. 
 The water of the more recent Citadel Green well contains 
 only 1 1 1.55 parts in f 00,000 of solid matter, likewise consisting 
 mainly of these two salts. The water is considered wholesome 
 as a drink, and, for washing, the presence of the carbonate of 
 soda makes it an excellent water ; the principal objection found 
 to it is that the carbonate of soda gives to rice, hominy and 
 other farinaceous articles cooked in it a light golden tinge, owing 
 to the action of the carbonate of soda on the starch in such arti- 
 cles. In the laundry, also, it cannot be used in the mixing of the 
 starch for the same reason. 
 
 It may be here noted that the quantity of water obtained from 
 an artesian well is often seriously diminished by the sinking of 
 other wells into the same water-bearing stratum. This has been 
 the experience in many localities. With reference to wells in the 
 neighborhood of London, Eng., De Ranee says : 
 
 "The outcrop of the lower London tertiaries is about lOO 
 feet above the Thames, whilst their depth below it varies from 
 200 to 300 f^et, the only notch in the rim of the basin being 
 the valley of the Thames at Deptford and Greenwich, where 
 the outcrop is lOO feet lower than the remainder of the margin 
 of the basin ; the sectional area of the depressed portion being 
 much less than the elevated portion, far less water can escape 
 than can be alDsorbed by the sands, which are practically water- 
 logged by the overlying, impermeable clay, through which borings 
 were carried to a depth of 80 to 140 feet at the beginning of the 
 century ; at that time the liberated water flowed up the bore- 
 holes, and rose permanently above the level of the Thames until 
 the supply was over-pumped, and it has fallen to 70 feet below 
 Trinity high-water mark. To supply the deficiency, most of the 
 artesian wells in London have been carried down to the chalk 
 beneath, to intercept the water which circulates freely in the fis- 
 sures and lines of joints. The level to which water will rise is 
 steadily decreasing." 
 
138 
 
 WATER SUPPLY. 
 
 Deep Wells. 
 
 In certain geological formations, the nature of which does not 
 admit of the construction of artesian wells proper, water may 
 often be obtained in large quantities by sinking shafts, in which 
 the water collects and from which it may be raised to the surface 
 by pumps. Horizontal tunnels may be carried from the shaft, at 
 one or more levels, so as to intercept the water flowing through 
 the fissures or along the planes of stratification. Sometimes, in- 
 deed, the water may be obtained solely by means of a horizontal 
 tunnel, the opening being made in the face of a bluff. Thus, 
 Dubuque, Iowa, is supplied from a tunnel or adit penetrating the 
 bluffs and extending for about a mile in length at a depth, from 
 the surface of the ground, of from lOO to 200 feet. 
 
 Deep wells are used as sources of public supply, to some 
 extent, but the greater number the world over are sunk or bored 
 for private establishments, notably for breweries. It is stated 
 that in the city of New York there are as many as 40 wells 
 on Manhattan Island, although some of them are not now in use. 
 Nearly one-half this number are owned by breweries. The wells 
 vary in depth from 26 to 2,000 feet. Eighteen are 500 feet or 
 more in depth. The diameter also varies, being from 2\ to 10 
 inches, although the majority are 6 or 6|^ inches. The capacity 
 of the wells ranges from 2,000 gallons to 126,000 gallons in 24 
 hours ; and the temperature of the water, so far as noted, is 
 between 50° and 59° F. As might be expected from the geolog- 
 ical formation of the island, the wells are, in most cases, bored 
 in gneiss and mica schist.* 
 
 In England, where deep wells are used to a considerable ex- 
 tent as sources of town supply, the water-bearing capacity of the 
 various geological formations, and the character of the water to 
 be obtained therefrom, has probably been studied more carefully 
 than elsewhere.f Of the water supply of Liverpool, 5,500,000 
 (imperial) gallons are daily pumped from the wells in the new red 
 sandstone, and London receives daily some 8,000,000 (imperial) 
 gallons from deep wells in the chalk. 
 
 * Sanitary Engineer, October 12, 1882. 
 
 f See Sixtli Report of Rivers Pollution Commission ; also, the various (annual) 
 reports of the Underground Water Committee of the British Association ; also, De 
 Ranee, Water Supply of England and Wales. 
 
DEEP WELLS. 
 
 139 
 
 F,ig. 26 shows a well which is sunk in the new red sandstone, 
 at Whiston, England.* Two wells, each 9 feet in diameter and 12 
 feet apart, were sunk to a depth of 135 feet and then continued 
 to a depth of 225 feet as a single well, 30 feet long and 9 feet 
 broad. The bottom of the well is 25 feet below mean sea level, 
 and when first sunk, supplied some 400,000 gallons in 24 hours, 
 with a depth of about 9 feet of water in the well. The manner 
 , in which the well cuts the strata of the sandstone is evident from 
 the figure, the water having a natural tendency to flow along the 
 planes of stratification toward the fault shown at the left, which 
 'presents a barrier to its farther progress. The works were sub- 
 sequently extended by sinking and boring the auxiliary well in 
 the right of the figure, and connecting the two wells by means 
 of a tunnel. The supply obtained from the combined wells, was, 
 in 1876, about 900,000 (imperial) gallons in 24 hours. 
 
 The capacity of a rock for storing and absorbing water varies 
 with its texture and character ; and when, after long continued 
 rains, the rock has become fully saturated, no more water can be 
 absorbed, and all additional supplies pass off as floods, as abso- 
 lutely as if the precipitation took place on an impermeable for- 
 mation. Many rocks thus contain, in their natural and undis- 
 turbed condition, water which was derived from the atmosphere 
 long ago, and in some cases the rocks may contain saline solutions 
 which have filled their pores from the time of their formation.f 
 When a well is sunk into,such a deposit, the water may be grad- 
 ually forced out by the pressure of the water accumulated in 
 other parts of the same stratum, or in communicating strata, but 
 it may take a very long time to exhaust the subterranean reser- 
 voir. In some cases, near the sea, there may be communication 
 with the ocean, which may thus produce the hydrostatic pressure, 
 but which may not contribute by its waters directly, or at least 
 not for a long time after the well is opened. Some deep wells 
 near the sea gradually become more brackish, probably from the 
 fact that the purer water which originally filled the pores of the 
 rocks, and perhaps subterranean reservoirs, is gradually exhausted, 
 and other water — in this case sea water — comes in to replace it. 
 
 Some idea of the vast amount of water stored below the sur- 
 
 * Proc. Inst. Civ. Eng. Gr. Britain, xlix (1877), p. 221. 
 
 t See Hunt's Chemical and Geological Essays, p. 104 and elsewhere. 
 
DEEP WELLS. 
 
 141 
 
 face of the ground may be obtained from the following extract 
 (De Ranee) : 
 
 " In the Thames and east coast district are not less than 4,000 
 square miles of pervious cretaceous rocks, receiving not less than 
 5 inches of rain annually, or a daily absorption of 800,000,000 
 gallons. It is readily understood with these figures, how the dry 
 weather flow of the Thames is kept up by chalk springs ; one- 
 fifth of the yield is sufificient for 4,000,000 people, and taking the 
 oolite supply, the total volume of water absorbed by underground 
 sources in the Thames and east coast river basins may be taken as 
 1,125,000,000 gallons — a supply equal to the wants of 22,000,000 
 people, or nearly that of the total inhabitants of England, sup- 
 posing that the whole of the 5 inches of rainfall absorbed could 
 be pumped up." 
 
 It is, of course, impossible to utilize all the water actually 
 contained in any rock. From a compact rock like chalk or lime-, 
 stone, a portion of the water is furnished by cracks and fissures, 
 and this -is readily given up ; another portion passes through the 
 rock itself, and although the water may be received and absorbed 
 with great rapidity, it is delivered with extreme slowness, and a 
 struggle is maintained, as it were, between capillarity and gravity. 
 
 Baldwin Latham * has called attention to the influence of the 
 barometric pressure on the volume of water discharged by springs 
 (or yielded by deep wells). When the barometer falls, the air 
 confined in the fissures of the rocks tends to expand and force out 
 the water, and the volume of the springs increases ; when there 
 is a rise in the barometer, there is a diminution of the flow. A 
 more or less marked coincidence between barometric changes 
 and variations in the amount of water discharged by mineral 
 springs was noticed long ago, and various explanations have 
 been offered to account for the phenomenon.! 
 
 When a well is opened in a water-bearing rock, the level of 
 the water, or plane of saturation, will 'be found to vary within 
 certain limits, being governed by the amount of rainfall absorbed. 
 This level, after extensive pumping, is artificially and locally 
 lowered, but, on the cessation of pumping, the original level is 
 restored by a sufilicient interval of rest, provided the volume 
 
 * Nature, September, 1881. 
 
 t See, for example, Alois Nowak : Ueber die barometrischen Ergiebigkeits- 
 Schwankungen der Quellen in AUgemeinem. Prag, 1880. 
 
142 WATER SUPPLY. 
 
 abstracted annually is not more than is supplied by the rainfall. 
 If the demand upon a deep well is greater than the supply re- 
 ceived directly or indirectly from the rainfall, the well will show 
 signs of exhaustion ; the supply may, however, be kept up by 
 deepening the well, that is, by taking the water from a lower 
 level. Many wells gradually furnish less and less water, because 
 in the beginning there was a quantity of water stored in the rock, 
 which has gradually become exhausted. On the other hand, 
 some wells furnish a supply increasing in abundance, owing to 
 the fact that the passages through which the water comes become 
 less obstructed and of larger size. Thus, according to the Rivers 
 Pollution Commission, every 1,000,000 (imp.) gallons of water 
 drawn from the chalk carries with it, in solution, on an average 
 I J tons of chalk through which it has percolated, causing an ad- 
 ditional storage room for 1 10 gallons of water ; so that the yield 
 of a well draining a given area in the chalk, other things being 
 equal, ought to gradually increase until the maximum limit of 
 permeability is reached. As a further example of the same 
 thing, it may be mentioned that, during the construction of the 
 tunnel at the Whiston water works, upward of 350 tons of sand 
 were in a few years washed from the fissures of the rocks, thus 
 increasing the storage capacity of the rock. 
 
 Characteristics and Examination of Deep-seated Water. 
 
 The questions of the amount of water to be obtained from 
 springs and deep wells, and of the probability of procuring water 
 by means of artesian wells in any given locality are questions for 
 the engineer and geologist. The fact that all such waters are 
 liable to contain an excess of mineral matter has been sufficiently 
 noticed : the chemical examination concerns itself mainly with 
 the amount and nature of the dissolved salts. These deep 
 waters are characterized, in general, by an absence of organic 
 matter, but that even deep wells are liable to pollution may be 
 easily realized by an inspection of Figure 26. It is very evident 
 that any polluting matters in the soil might easily find their way 
 into the well, being carried downward by the water passing 
 along the planes of stratification. In the case of the water works 
 at Whiston, all the wells in the immediate neighborhood were 
 affected, most of them losing their water altogether. This in- 
 
DEEP WELLS. 
 
 143 
 
 fluence was felt at least i^ miles to the southeast and more than 
 a mile to the south. Where the water flows underground 
 through cracks and fissures the polluting substances do not have 
 the opportunity to become oxidized and harmless, as when the 
 water passes slowly through a gravel deposit. Mr. Baldwin 
 Latham has connected the periodic outbreaks of fever in the 
 parish of Croydon, England, with the intermittent appearance 
 of springs called the Bourne. The water which is reabsorbed by 
 the chalk lower down, is supposed to carry the objectionable 
 substances to the wells in the center of the old town. When 
 the springs are low and the Bourne begins to run, after a sudden 
 and copious rainfall, the water line under the town is elevated 
 and an outbreak of enteric fever results. Even with artesian 
 wells there is not perfect security, "for many such wells — espe- 
 cially when first opened — throw out fragments of vegetable sub- 
 stances, and even living fish and other small aquatic animals, 
 showing that they must have a more or less direct communica- 
 tion with the surface. 
 
 Signs of pollution in such waters must be sought mainly in 
 the " organic matter " as variously determined : chlorides are 
 often present in considerable amount, but are not evidence of 
 impurity ; nitrogen in the form of nitrites and nitrates is also 
 often present in notable quantity in water of deep wells, espe- 
 cially in the chalk, and is no sign of contamination ; ammonia 
 may also be allowed in amounts which would be suspicious in 
 shallow wells. 
 
 While the artesian wells often furnish water the temperature 
 of which is objectionably high, the water of many springs and of 
 ordinary deep wells is usually of nearly uniform and of compar- 
 atively low temperature. In fact, the sinking of deep wells in 
 connection with breweries is partly due to the fancied necessity 
 for hard water in brewing certain kinds of beer, and partly to 
 furnish an abundance of cold water for cooling purposes. 
 
 It is a curious fact that the hard water from springs and depp 
 wells, though clear and bright when first obtained, becomes 
 covered with a confervoid growth when exposed to the sunlight 
 in open reservoirs, and the tubes of some artesian wells become 
 lined with a growth of algae. If it is necessary to store the 
 water from deep wells, this should be done in covered reservoirs ; 
 this is, of course, desirable also as a means of avoiding elevation 
 
144 
 
 WATER SUPPLY. 
 
 of temperature in summer if the water remains in the reservoir 
 for any considerable time. 
 
 Table XIX contains some details with reference to artesian 
 
 TABLE XIX.— Examination of Artesian and Deep Wells. 
 
 Locality. 
 
 Artesian Wells. 
 
 Crenelle, Paris 
 
 Passy, Paris 
 
 Boston, Mass 
 
 Chicago, 111 
 
 Louisville, Ky., Dupont's Well 
 
 St. Louis, Mo 
 
 " Mo., Asylum Well. . 
 
 Charleston, S. C, Wentworth Street 
 " " Citadel Green, 
 
 " " ChlsholmMill. 
 
 Coosaw, S . C 
 
 Deep Wells. 
 
 Birkenhead, Eng- 
 
 Birmingham, Eng 
 
 " Another 
 
 Bradford, Eng 
 
 Brighton, Eng , 
 
 Liverpool , Bootle Well 
 
 London, Trafalgar Square 
 
 Jersey City, N. J., Secaucus Works 
 Newark, N. J., Celluloid Works. . . . 
 
 " " Lister Bros 
 
 Paterson, N. J., Burton Brewing Co 
 
 ^ 
 
 g 
 
 
 
 
 Bl U 
 
 
 
 u fc s 
 
 
 Q 
 
 p 
 
 H 
 
 X 
 
 H 
 
 i,8o6 
 
 27 
 
 
 14.2 
 
 1,914 
 
 28 
 
 
 
 14. 1 
 
 1,75^ 
 
 
 
 
 1B78.7 
 
 700 
 
 ■ 14 
 
 
 
 
 2,086 
 
 24. s 
 
 
 
 1570.0 
 
 2,199 
 
 23 
 
 
 
 879.1 
 
 3.843 
 
 40. s 
 
 
 
 
 1,200 
 
 30.7 
 
 
 
 273.7 
 
 1,970 
 
 37-5 
 
 
 
 111.6 
 
 425 
 
 
 
 
 369-7 
 
 760 
 
 
 
 
 82.7 
 
 527 
 
 
 5-Z 
 
 14-2 
 
 300 
 
 10.2 
 
 15. B 
 
 -3>-3 
 
 400 
 
 10.8 
 
 IS. I 
 
 19-3 
 
 36= 
 
 12.8 
 
 14. 1 
 
 SS-4 
 
 1,28s 
 
 9-9 
 
 4.4 
 
 35-4 
 
 312 
 
 10.4 
 
 I2.6 
 
 34-4 
 
 383 
 
 
 5-9 
 
 83.4 
 
 600 
 
 
 
 "7-7 
 
 250 
 
 .... 
 
 
 213.0 
 
 615 
 
 13.0 
 
 
 262.0 
 
 200 
 
 
 
 
 20.6 
 
 Authority. 
 
 Peligot, 3857. 
 
 Poggiale & Lambert, 1862. 
 
 J. W, Merrick. 
 
 J. Lawrence Smith. 
 A. Litton, M.D. 
 C. C. Broadhead. 
 C. U. Shepard,Jr. 
 S. T. Robinson, Jr. 
 Wm, Robertson. 
 F. F. Cliisholm. 
 
 Rivers Pollution Com. 
 
 G. H. Cook, Geol. Rep. 
 
 and deep wells in various localities. Although there are so 
 many of these wells in the United States, the author has found 
 it extremely diiificult to obtain reliable information with reference 
 to the chemical character of the water. The depth of the well, 
 whether the water is palatable or undrinkable, how it acts to- 
 ward soap and in steam boilers — these observations, which do 
 not require the aid of an expert or involve additional expense, 
 usually complete the stock of available information. Table XX 
 contains the results of more complete analyses obtained by the 
 Rivers Pollution Commission of Great Britain from the examina- 
 tion of a large number of deep-well waters from various geolo- 
 gical formations. In the same table also — for convenience of 
 arrangement — are inserted the results obtained by the same 
 commission in the examination of unpolluted waters of various 
 descriptions and from many different localities in Great Britain. 
 Table XXI contains results derived from the examination of the 
 water from wells and springs in the different geological forma- 
 tions in Bohemia,* the total number of analyses being about 125. 
 
 * B^lohoubek : Ueber den Einfluss der geologischen Verhaltnisse auf die chemische 
 Beschaffenheit des Quell- und Brunnenwassers. Pph. 8vo, pp. 46. Prag, 1880. 
 
EXAMINATION OF VARIOUS WATERS. 
 
 I4S 
 
 TABLE XX.^ExAMiNATiON of'Deep-Well Waters and of Unpolluted 
 Waters from various Sources. 
 
 [Results expressed in Parts in icxj,ooo.j 
 
 Geological For- 
 mation. 
 
 Deep Wells. 
 
 Devonian Rocks and 
 Millstone Grit. . . 
 
 Coal Measures 
 
 Magnesian Lime 
 stone 
 
 New Red S a n d - 
 stone 
 
 Lias 
 
 Oolites 
 
 Hast^ n g s Sand, 
 Green Sand, and 
 Weald Clay . . . 
 
 Chalk 
 
 Chalk beneath Lon- 
 don Clay 
 
 Thanet Sand and 
 Drift 
 
 Unpolluted Waters. 
 
 Class L 
 
 Rain Water 
 
 Class H. 
 
 Upland Surface 
 
 Water 
 
 Class III. 
 
 Deep Well Water. 
 Class IV.. 
 
 Spring Water 
 
 32.68 
 83.10: 
 
 61.14 
 
 30.6^ 
 70.98 
 33.60 
 
 45.20 
 36.88 
 
 78 09 
 
 53- 
 
 2.95 
 
 9.67 
 43.78 
 
 0.068 
 0.119 
 
 0.076 
 
 o 036 
 0,146 
 0.037 
 
 0.068 
 0.050 
 
 0.093 
 
 0,113 
 
 0.070 
 
 0.322 
 0.061 
 0.056 
 
 0.012 
 0:034 
 
 0.0T4 
 0.027 
 0,010 
 
 0,014 
 0.017 
 
 0.005 
 0.044 
 
 0.003 
 0,001 
 0.022 
 
 0,016 
 o .001 
 
 0.048 
 
 0.020 0.072 
 
 0.032 
 
 0.018 0.012 
 o 001 
 
 zia 
 
 < o 
 
 O H 
 
 0.294 0.310 
 0.207 0.278 
 
 426 1.456 
 
 o 717 
 
 0.389 
 
 0.625 
 
 o.ig6 
 0.610 
 
 0.06S 
 
 G.116 
 
 0.003 
 
 0.00^ 
 0.495 
 o 383 
 
 o 734 
 0.417 
 0.654 
 
 0.723 
 0.628 
 
 ° 135 
 0.202 
 
 o 042 
 
 0.042 
 0.522 
 0.396 
 
 
 
 
 
 1 
 
 = 2 
 
 
 Hardness. | 
 
 US 
 
 
 
 
 
 
 
 
 
 
 < H 
 
 
 
 
 
 3; z 
 
 
 
 
 
 So 
 
 
 
 
 
 inU 
 
 
 fc- 
 
 ^ 
 
 
 D < 
 
 2 
 
 
 
 S^z 
 
 
 
 
 
 S3 2 
 
 s 
 
 I 
 
 I 
 
 a 
 
 ;§ 
 
 a. 
 
 u 
 
 H 
 
 Ph 
 
 E- 
 
 2,671 
 
 2.70 
 
 8.8 
 
 8.6 
 
 17-4 
 
 2>243 
 
 18.05 
 
 15.1 
 
 20.6 
 
 35-7 
 
 13.937 
 
 4.31 
 
 16.9 
 
 26.9 
 
 43.8 
 
 6,89s 
 
 2 94 
 
 7-4 
 
 JO. 5 
 
 17.9 
 
 3.730 
 
 4.42 
 
 21. 9 
 
 8.2 
 
 30.1 
 
 6,118 
 
 2.69 
 
 I3.» 
 
 6.8 
 
 20,6 
 
 1,864 
 
 5.38 
 
 16.8 
 
 10.5 
 
 27.3 
 
 5,8-Ji 
 
 2.76 
 
 21 2 
 
 6.5 
 
 27.7 
 
 797 
 
 15.02 
 
 9.7 
 
 8.7 
 
 18.4 
 
 '.SI 7 
 
 6.32 
 
 14.4 
 
 7.6 
 
 22.0 
 
 42 
 
 0.82 
 
 0.4 
 
 o.S 
 
 0.3 
 
 10 
 
 '.13 
 
 1-5 
 
 4-3 
 
 5.4 
 
 4,743 
 
 5. II 
 
 15.8 
 
 9.2 
 
 25 
 
 3.S95 
 
 2.49 
 
 II. 
 
 7-5 
 
 18.5 
 
 66 
 13 
 4 
 
 19s 
 157 
 
 TABLE XXI. — Bohemian Well and Spring Waters. 
 [Results expressed in Parts in 100,000.] 
 
 Geological 
 Formation. 
 
 Azoic 
 
 Huronian .... 
 (( 
 
 Silurian 
 
 X 
 
 Carboniferous . 
 
 Permian 
 
 Chalk..".'.'.'.".'. 
 
 Neogen 
 
 Diluvium and 
 Alluvium 
 
 }.... 
 
 Results. 
 
 Max. and Min. 
 
 Usual Limits. 
 Max. and Min. 
 
 Usual Limits. 
 Max. and Min. 
 
 Usual Limits. 
 Max. and Min. 
 
 Usual Limits. 
 Max. and Min. 
 
 Usual Limits. 
 Max. and Min. 
 
 Usual Limits. 
 
 Usual Limits. 
 
 Usual Limits. 
 
 Total 
 Solids. 
 
 2.8- 2S.0 
 
 4.6- 14.3 
 
 19.8- 55.0 
 
 30.5- 55 o 
 
 14.5-150-2 
 66.0-120.0 
 38.5-374.0 
 
 4.5- 29.7 
 6.5- g.o 
 4.0-iog.o 
 
 21.5- 75-7 
 
 37.0- 49.7 
 
 19.1- 75.5 
 
 Sulphuric 
 Acid iSOj). 
 
 0.2- 2.4 
 
 1.3- 5-9 
 
 2.8- 48.6 
 
 2.4-IOI.6 
 
 O.I- 1.5 
 
 0.3- 28.3 
 
 0.7- 6.8 
 
 0.6- 5.7 
 
 Chlorine. Hardness.* 
 
 O.I- 0.8 ' 1.3- 6.8 
 
 ■ 2.5- 4.9 
 
 0.9- 2.4 , 7-7- 13-9 
 
 0.5-12.8 4 7- 57-3 
 
 22.6- 45.4 
 
 0.8-25.2 
 
 10.3-124.9 
 
 0.1- 1.3 19-9- 22.8 
 
 1-3- 13.3 
 
 0.1- 6.8 2.1- 3-5 
 
 1.3- 27.1 
 
 1.8- 4.6 19.2- 21.3 
 
 1.8- 2.3 ' 7.9^ 24.5 
 
 * The hardness is in German degrees. 
 
CHAPTER VIII. 
 
 ARTIFICIAL IMPROVEMENT OF NATURAL WATER. 
 
 Nature sometimes furnishes water which leaves nothing to 
 be desired in respect to quaHty, but, unfortunately, the best 
 water — such as is sometimes derived from springs and not un- 
 frequently from the ground water— is apt to be limited in 
 amount, and, very generally, a supply suflScient in quantity can 
 be procured only from a source possessing some undesirable 
 qualities. Although a water which is polluted to any consid- 
 erable extent by sewage is not capable of purification by any 
 process practicable on the large scale, so that it can safely be 
 used for domestic supply, the character of many natural waters 
 may be sensibly improved by proper treatment. Again, there 
 are circumstances under which it becomes absolutely necessary 
 to use a bad water or none at all, and in such cases purification 
 must be accomplished if possible. 
 
 We shall consider various methods of improving potable 
 water under the following heads : 
 
 1. Sedimentation and storage. 
 
 2. Filtration (on the large and on the household scale). 
 
 3. Clark's process for softening hard water. 
 
 4. Other (chemical) processes. 
 
 5. Distillation. 
 
 Sedimentation and Storage. 
 
 When a stream or other body of surface water is used as a 
 source of supply, the best way — speaking solely from a sanitary 
 point of view — is to pump directly from the source into the dis- 
 tribution without the intervention of settling basins and reser- 
 voirs. This is not, however, generally practicable. In some 
 cases the general or occasional turbid character of a stream 
 renders sedimentation necessary, and in other cases storage 
 basins are necessary in order that the stream may furnish a suf- 
 ficient quantity of water during the dry season. 
 
SEDIMENTATION. 
 
 147 
 
 Sedimentation. — The particles of suspended matter which 
 render a stream turbid or muddy, regularly or in times of flood, 
 are of a greater specific gravity than the water itself, and settle 
 out more or less completely if the water be allowed to stand 
 quiet for a time. Lakes and ponds are natural settling basins, 
 and have this advantage over running streams, that they are 
 much less liable to be rendered turbid by freshets ; they are not 
 however, usually entirely free from floating particles, but the 
 suspended matter in this case cannot be removed except by 
 filtration. 
 
 Some figures have been already given in Table VII to show 
 the various amounts of suspended matter in certain rivers. It 
 is well known that the water of many of the rivers in this coun- 
 try, especially in the West, is not, in its natural condition, 
 acceptable, if indeed it can be regarded as at all suitable for 
 use. It is said* that the sediment in the Mississippi (or rather 
 the Missouri.) at St. Louis amounts at times to 1.8 per cent of 
 the bulk of the water. About 94.5 per cent of the sediment is 
 deposited within 24 hours in still water at ordinary stages of the 
 river, but for two months in the year, during floods, there is so 
 much fine sediment that no amount of settling will clarify the 
 water. 
 
 It is desirable to remove the suspended matter, not simply 
 on Esthetic grounds, because it renders the water less acceptable 
 to the eye, but also because particles of gritty mineral matter, or 
 even of clay, often cause diarrhoea, especially in the case of per- 
 sons not in the habit of using the water, to which many persons 
 do become accustomed by use. Where thorough filtration is for 
 any reason impracticable, sedimentation serves a useful purpose. 
 
 There are now at St. Louis four settling basins, 600 x 278 feet, 
 and 19 feet deep. The floors are paved with brick on edge and 
 slope toward the center and the river side. The basins are used 
 alternately, one being drawn from, one filling and two settling, 
 or one settling and one being cleaned. The sediment is removed 
 from each basin about once in four months, some 16 inches 
 being collected in that time. The sediment is floated off with a 
 stream of water, at a cost of from 0.84 cent (1879) to 1.05 (1878) 
 per cubic yard. 
 
 Engineering News, 1881, p. 142. 
 
148 
 
 WATER SUPPLY. 
 
 As an accompaniment of a scheme of filtration, settling basins 
 are in many cases essential. Where a stream is subject to 
 sudden and transient periods of turbidity, besides serving in their 
 proper capacity, they may also serve as storage basins and make 
 it possible to avoid altogether taking water from the stream 
 when it is at its worst. 
 
 In following the variations of a turbid water, or in tracing the 
 progress of sedimentation, estimation of the amount of turbidity 
 is usually made by simply looking through a certain depth of the 
 water in a tube or other glass vessel. Several attempts have 
 been made to reach more accurate results, and Grahn and Sal- 
 bach'* have proposed to apply the principles of photometry to 
 the solution of the problem. 
 
 Fig. 27. — salbach's photometer, elevation, (from fischer.) 
 
 Fig. 28. — salbach's photometer, plan, (from fischer.) 
 
 Figures 27 and 28 represent Salbach's arrangement of the 
 photometer. P carries a disk of paper on which a spot has been 
 made transparent by means of oil or stearine ; at the beginning 
 P is set at the zero point of the scale, and the burners, B B, reg- 
 ulated so as to give an equal amount of light. (7 is a leaden box 
 the glass sides of which are exactly parallel, and 100 millimeters 
 apart. When C is filled with turbid water a portion of the light 
 
 Journal fUr Gasbeleuchtung und Wasserversorgung, xx (1877), p. 545. 
 
STORAGE. 
 
 149 
 
 given by the right-hand burner is cut off, and the screen P must 
 be moved nearer to this weakened source of hght in order 
 that the illumination on both sides may be equal. From the 
 distance through which the screen must be moved, it is possible 
 to calculate the loss of light, and thus to obtain an expression 
 for the amount of turbidity of the water. 
 
 Storage. — Some of the disadvantages to which stored water is 
 subject in ponds and impounding reservoirs have already been 
 indicated — such, for instance, as undesirable increase of temper- 
 ature, growth of noxious algae, etc. On the other hand, there 
 are certain advantages in storing surface waters in clean deep 
 basins exposed to the sun and air. When strongly colored sur- 
 face waters are thus stored, the dissolved organic matter under- 
 goes chemical change, and a portion of it, being removed from 
 solution, settles out as a sediment ; at the same time the color 
 becomes less marked. 
 
 In order to compare and estimate the depth of color of vari- 
 ous waters, or of the same water at various times, an instru- 
 ment was invented and used by a medical commission which 
 investigated a number of sources of water supply proposed for 
 the city of Boston.* "The instrument consists of two tubes, B 
 and D (Fig. 29), sliding water-tight one within the; other, the 
 lower end of each tube being 
 closed with a disk of plate 
 glass. Into the large tube, B, 
 just above the plate glass disk, 
 is inserted a small piece of 
 tubing which terminates in a 
 funnel-shaped receiver, A. 
 Water poured into this re- 
 ceiver will therefore pass into 
 the space between the two 
 glass disks, entirely filling the 
 outer tube when the inner 
 tube is withdrawn, and again ^'^°- ^9- 
 
 returning to the receiver when the inner tube is pushed down so 
 that the glass disks come in contact with each other. Through 
 
 * Report of the Medical Commission upon the Sanitary Qualities of the Sudbury, 
 Mystic, Shawshine and Charles River Waters. City Document, No. 102. Boston, 
 1874. 
 
150 WATER SUPPLY. 
 
 an opening near the upper end of the smaller tube, D, is inserted 
 one end of a rhombic prism, E, in which total internal reflection 
 takes place twice. 
 
 " This prism extends half way across the inner tube, D, so 
 that an eye, looking through the eye-piece, G, sees the field of 
 vision nearly half filled by the surface of the prism. This ap- 
 pearance is represented at /. The eye-piece, G, contains a single 
 lens, which is focused upon the upper surface of the prism. The 
 position and angles of the prism are such that a ray of light out- 
 side of, and parallel to, the tube B, is reflected, first directly into 
 the tube D, and then parallel to its axis, thus emerging from the 
 prism and entering the eye-piece alongside of the rays of light 
 which have passed through the two plate glass disks. It will 
 thus be seen that the conditions for comparing the color and 
 intensity of these two sources of light are as favorable as pos- 
 sible. A piece of white card, C, fastened at the lower end of the 
 larger tube, throws a uniform white light through the tubes 
 B and D, and also along the outside of the tube B, into the 
 prism E. 
 
 " In using this instrument, a piece of brownish-yellow glass, F, 
 is placed in front of the prism E, and the water, whose color is 
 to be determined, is poured into the receiver A. The inner tube 
 is then withdrawn until the column of water between the two 
 glass disks is sufficiently long to give to the light passing through 
 it a color equal to that imparted by the colored glass F to the 
 light passing through the prism E. The length of this column 
 of water, which will, of course, vary inversely with the depth of 
 color, can be determined by means of the scale on the inner 
 tube D. By this means, the relative intensity of color of various 
 specimens of water may be determined with considerable ac- 
 curacy." , 
 
 Aeration. — Although a water colored by vegetable matter 
 loses color when exposed to sunlight in closed vessels, it is prob- 
 able that the changes which take place in storage basins are due 
 partly to the action of the air. There is no doubt that the pas- 
 sage of the water over natural falls or artificial dams tends to its 
 improvement, or that the stagnation of water in " dead ends," or 
 in small reservoirs without circulation, tends to its deterioration. 
 In the larger reservoirs and ponds the winds play an important 
 part in agitating and aerating the water, but it is very doubtful 
 
FILTRATION. 1 5 i 
 
 if any scheme for artificially aerating the water would accom- 
 plish enough to pay for the outlay involved. 
 
 Filtration. 
 
 The filtration of water on the large scale has been practised 
 in England and on the continent of Europe for many years, and 
 has become very general in cases where the supply is taken from 
 streams or ponds. From statistics which were laid before the 
 Diisseldorf meeting of the German Public Health Association 
 (in 1876) by Engineer Grahn,* it would seem that in Germany, 
 since 1858, there has been no town of considerable size supplied 
 with unfiltered river water, while the increase with reference to 
 other sources of supply may be seen from the following data : 
 
 TOTAL NUMBER OF INHABITANTS IN 8o TOWNS OF GERMANY, GERMAN-AUSTRIA, 
 
 AND SWITZERLAND 
 
 Supplied with 
 
 1858. 
 
 1876 
 
 
 460,000 
 
 1,060,000 
 
 25,000 
 
 45,000 
 
 460,000 
 1,697,000 
 1,519.000 
 1,719,000 
 
 Filtered river water 
 
 Spring and ground water (by gravitation) 
 
 
 
 In the United States very little has yet been done in the way 
 of systematic filtration on the large scale, although, in some lo- 
 calities, attempts have been made to improve the condition of 
 the water by intercepting some of the suspended matter. 
 
 Filter beds, as usually constructed, are water-tight basins 
 some ten feet or more in depth, the sides built of masonry, and 
 the bottom puddled or made of concrete, or paved with brick 
 and cemented. The area may be from 20,000 to 50,000, or in 
 some cases eiven 150,000 square feet. In building up the filter- 
 ing bed, provision is first made for the ready collection of the 
 water by constructing upon the floor of the basin drains or 
 channel-ways of stone or brick laid dry ; then follows a layer of 
 broken stone, the fragments being three or four inches in diame- 
 ter. This is succeeded by gravel screened so as to be of uniform 
 size, a layer of coarse being followed by one or more layers of 
 finer material ; upon the gravel rests sand, likewise separated 
 
 * See the Deutsche Vierteljahrsschr. fur offentl. Gesundheitspflege, ix (1877), p. 
 108. 
 
152 WATER SUPPLY. 
 
 into layers of uniform size. The exact thickness of the different 
 layers, and the extent to which the separation into the differe/it 
 sizes is carried, are subject, of course, to considerable variation. 
 
 The water stands several feet deep over the surface of me 
 sand, and is allowed to flow down through the filter at such rate 
 as experience shows to be most advantageous. Naturally, when 
 the sand is clean, a greater quantity of water can be passed lin a 
 given time than when the sand has become clogged ; praqtice 
 differs as to the maximum rate, but it is seldom over six inches, 
 vertically, per hour, and often less. At the rate mentioned, each 
 square foot of surface would deliver I2 cubic feet (or 89^ United 
 States gallons) per day. 
 
 When the beds become clogged so as no longer to filter with 
 sufficient rapidity, the water is drawn out from the beds, and the 
 upper layer of sand, for a depth of one-half or three-quarters of 
 an inch, is removed. When by successive parings the thickness 
 of the sand has been considerably reduced, that which has been 
 removed is washed and replaced so as to restore the original 
 thickness, the waste of washing being made up with fresh sand. 
 
 Principles of Sand Filtration. 
 
 Having thus stated briefly the main features of ordinary- 
 sand filtration, we will proceed to discuss the principles which 
 govern filtration in general, and afterward consider certain 
 special points with reference to the successful carrying out of 
 the process. For this purpose we shall consider the " impuri- 
 ties " of ordinary water as divided into three classes : first, the 
 suspended matters, whether of mineral, animal or vegetable 
 origin ; second, the dissolved mineral or saline matters ; third, 
 the dissolved organic matters. 
 
 The action which takes place when an ordinary water is 
 passed through a sand filter is threefold. In the first place, the 
 most obvious action is the arresting of suspended particles of 
 solid matter which are too large to pass through the pores of 
 the filter. The second action partakes something of the charac- 
 ter of sedimentation, and may be well illustrated by the follow- 
 ing experiment: 
 
 Take two jars of equal size, and fill one of them with frag- 
 ments of broken rock as large as half a fist, or with very coarse 
 gravel, arranging the material so that a syphon can be inserted 
 
PRINCIPLES OF FILTRATION. 
 
 153 
 
 Fig. 30. 
 
 as shown in the figure. Prepare now a quantity of a turbid hquid 
 by stirring up 
 some fine clay in 
 water and allow- 
 ing the coarser 
 particles to sub- 
 side. With this 
 turbid water fill 
 both jars and al- 
 low them to 
 stand for twelve 
 or fifteen hours. 
 Then, by means 
 of syphons reach- 
 ing to the same 
 depth in the 
 
 jars, carefully remove a quantity of water from each. It will be 
 found that the water from the jar containing the broken stone is 
 perceptibly clearer than the other. 
 
 The same thing may be shown by allowing a turbid water to 
 flow very slowly through a trough filled with broken stone. In 
 these cases the interstices between the fragments are so many set- 
 tling chambers, as it were, and the particles of the clay deposit 
 not only upon the floor of these chambers, but also on the sides 
 and roof, being drawn thereto by a sort of attraction. In the 
 case of a sand filter, the interstices are small and very numerous. 
 
 The third sort of action which takes place in the porous 
 material of a filter is the removal of substances which are actually 
 dissolved in the water. As far as the mineral or saline matters 
 are concerned, this action is trifling, although not inappreciable 
 with certain filtering media : with sand filters we can say that 
 there is, practically, no effect on the dissolved mineral matter, 
 unless there is opportunity for a chemical change to take place. 
 Thus, it seems to be well attested, that a hard water containing 
 bicarbonate of lime may deposit carbonate of lime in the filter, 
 owing to the escape of carbonic acid.* Sometimes the amount 
 
 * See, for example, Lefort, Chimie Hydrologique, pp. 165, 200. It has also 
 been shown by Schloesing (see Assainissement de la Seine, 2ieme partie, Enquete, p. 
 191), that at some depth in soil which had been irrigated with sewage, there were 
 formed crystals of carbonate of lime, owing to the escape of carbonic acid from the 
 sewage water, which contained a small proportion of bicarbonate of lime in solution. 
 
154 WATER SUPPLY. 
 
 of mineral matter may be greater in the filtered than in the un- 
 filtered water, if the material of the bed, the gravel and stones, 
 contain, as they often do, soluble ingredients. 
 
 With reference to the dissolved organic substances, it may 
 be said that a small but appreciable amount may be removed by 
 a well-conducted sand filtration ; the action may be explained in 
 two ways. In the first place, most porous substances possess 
 the power of removing certain kinds of organic matter by some- 
 thing which' may be called adhesion. The absorptive power for 
 any substance is limited and soon reached, and the substance 
 thus removed may by appropriate means be again brought into 
 solution. Quartz sand, as we should infer, possesses the power 
 to a slight degree only. In the second place, dissolved organic 
 matter is removed in the sand 'filter by oxidation. The sub- 
 stance is actually burned more or less completely, in part by the 
 oxygen held in solution in the water, and in part by the air en- 
 tangled in the interstices of the sand. Although in filling the 
 beds with water, great care is taken to displace the air gradually, 
 and as completely as possible, there must always some remain in 
 the concavities of the individual grains of sand and otherwise 
 entangled. The extent of the action of a sand filter in this 
 direction depends not only on the fineness of the filtering 
 medium, and the rate at which the filtration takes place, but also 
 and in considerable measure upon the frequency with which the 
 filter is cleansed. The cleansing of the filter not only removes 
 the accumulation of organic matter, which, if allowed to remain, 
 would tend to injure the water, but also involves the aeration of 
 the sand, at least to a considerable depth. 
 
 Details of Practice. 
 
 We now proceed to some more particular details of the prac- 
 tice of sand filtration. The quality of the sand employed is by 
 no means a matter of indifference, and in the case of waters 
 which usually or occasionally carry finely divided clay in suspen- 
 sion, a great deal depends upon having proper sand and a slow 
 rate of filtration. It may be said, in general, that the sand em- 
 ployed should be made up mainly of grains from ^ of a milli- 
 meter to I millimeter in diameter, and the more uniform the size 
 
DETAILS OF PRACTICE. I 55 
 
 of the grains the better.* A considerable proportion of larger 
 grains does no harm, but the smaller particles should be washed 
 out before the sand is used ; as a rule, sand that has been used 
 and washed is better than fresh sand. As the fineness of the 
 sand increases, its efificiency as a filter increases, but the dif^culty 
 and cost of filtration increase likewise, and more frequent cleans- 
 ing is necessary. To obtain the best and most economical results 
 with a given water, special experiments should be made with 
 reference to the sand best fitted for that water — although, to be 
 sure, it is not always possible to command that which would be 
 absolutely the best. 
 
 While it is, in general, true that the upper layer of sand does 
 most of the work in intercepting the various floating matters in 
 the water, it does not do the whole under the conditions which 
 occur in ordinary practice. Examination shows that the sand is 
 somewhat affected to a greater depth, and it may occasionally be 
 necessary to renew all the sand. The very fact, which will 
 appear presently, that in all actual works there are times when 
 the water is imperfectly clarified, shows that the interior of a 
 sand filter must in time become more or less foul. 
 
 It may perhaps be asked, why, if the work is practically done 
 by the first few inches of sand, it is necessary to bestow such 
 care on the construction of the beds, and on the arrangement of 
 the materials employed. 
 
 In the first place, it is a well-recognized fact, that the worst 
 possible filter is one in which the portions of material of different 
 sizes are indiscriminately mixed. " The different degrees of fine- 
 ness in the materials beneath the sand, and their several thick- 
 nesses, were intended first to prevent the fine sand from following 
 the water downward into the drains, and next to insure the 
 presence of such a body of clean water below the surface of the 
 filter as would penetrate the numerous joints and openings of the 
 drains, and keep them full, without creating anywhere currents 
 or veins of water of any perceptible difference of velocity. 
 
 " With the drains much nearer to the body of the sand, it 
 will be understood that the tendency of the water would be to 
 flow through the filtering material more rapidly just over the 
 
 * Mr. Charles Greaves, of the East London Works, says "sand that would go 
 through a screen consisting of 32 or 33 No. 10 wires in 6 inches," is the best, accord- 
 ing to English experience. 
 
156 
 
 WATER SUPPLY. 
 
 pipe than at five feet on either side of it. The distance through 
 which it had to travel might be so short as to induce its concen- 
 tration. The low velocity at which the water flows through the 
 filter, the uniformity of fineness in the sand, and the distance of 
 the collecting drains from its surface, all work together to pro- 
 duce that regularity of action over the entire filter bed upon 
 which its perfection depends." * 
 
 The rate of filtration must be determined by the character of 
 the water and the condition of the filters. The maximum rate 
 given on page 152 as 89^ U. S. gallons per square foot in 24 
 hours would be equivalent to about 3^ cubic meters per square 
 meter of surface. The practice of most works falls considerably 
 below this as an average rate ; thus at Altona, wherei to be sure, 
 the constantly turbid water of the Elbe is filtered, the average 
 rate is only 1.5 cubic meters per square meter of surface in 24 
 hours. It is stated that the sand here employed is coarser than 
 necessary, and that with somewhat finer sand a more rapid rate 
 of filtration would be possible, f 
 
 There is considerable difference in practice as to the depth of 
 water kept upon the surface of the beds and the head under 
 which filtration takes place. Moreover, the head under which 
 the water is filtered varies at any works according to the condi- 
 tion of the sand. The clear-water 
 well is generally so arranged that 
 the height of water in it can be 
 lowered at pleasure ; and the head 
 under which the water is filtered 
 is the difference between the level 
 in the bed and in the clear-water 
 well, as may be seen, in the accom- 
 panying cut, where the head is 
 measured by the distance between 
 r a and b. While the beds are clean, 
 a difference of from 9 to 12 inches 
 sufifices to cause a proper rate of 
 Fig. 31- flow ; as they become clogged a 
 
 much greater pressure is required, but it is not desirable to in- 
 
 * Kirkwood, Filtration of River Waters, p. 10. 
 
 ■f Samuelson, Nachschrift to German edition of Kirltwood's Report. Hamburg, 
 
 1876. 
 
DETAILS OF PRACTICE. 
 
 157 
 
 crease the pressure to too great an extent, as the sand is thereby 
 fouled to a greater depth and compacted more than is desirable. 
 
 The frequency with which it is necessary to cleanse the beds 
 depends upon circumstances. In the worst stages of the English 
 rivers a filter bed has to be cleaned once a week, rarely oftener. 
 When the rivers are free from turbidity, cleansing may not be 
 necessary more than once a month, or in some cases once in two 
 months. The general method of cleaning has already been indi- 
 cated. In some places the practice is different. At Zurich, in 
 Switzerland, it is the custom to clean the beds by forcing the 
 water in a reverse direction through the filters. Workmen, clad 
 in rubber clothing, then stir up the upper surface of the sand 
 with forks, and the collected slime is washed off and floated 
 away. This requires an abundance of water, and the ability to 
 command the requisite pressure. One would suppose also that 
 it would involve some danger of disturbing the arrangement of 
 the filtering material. 
 
 Certain recently published details of the manner in which the 
 filtration works at Berlin* are conducted, are of considerable in- 
 terest, as the condition of things is similar to that which exists in 
 some of the surface waters of our Eastern States. The water of 
 the Spree, besides being somewhat polluted by sewage and being 
 in a constant roily condition, possesses, especially in time of 
 flood, a deep brownish-yellow color, and, at times, a peculiar 
 pondy taste due to vegetable extractive matter. Moreover, from 
 spring until fall, a more or less copious growth of algce adds to 
 the disagreeable character of the water, similar to those described 
 and figured on page 86. 
 
 The filter beds are construced on the English model and are 
 eleven in number — three covered and eight uncovered — having a 
 total area of 37,000 square meters. The filtration is carried on 
 at a very slow rate. The water is used, of course, in varying 
 quantities from hour to hour, and, on account of the small size 
 of the clear-water reservoir, a constant rate of filtration is impos- 
 sible ; the maximum rate is, however, not over o.i meter down- 
 ward per hour. For the greater part of the time i square meter 
 
 * Mittheilungen uber natUrliche und kUnstliche Sandfiltration. Nach Betriebs- 
 resultaten der Berliner Wasserwerke vor dem Stralauer Thor, bearbeitet von C. Piefke, 
 Eetriebs-Ingenieur . 8vo, pp. 75. Berlin, 1881. A review and abstract of this 
 pamphlet appeared in the Journal of the Franklin Institute, December, 1881. 
 
158 WATER SUPPLY. 
 
 of sand surface is not required to furnish much more than one 
 cubic meter of water in twenty-four hours. This would be at 
 the rate of only 24^ U. S. gallons per square foot, and very much 
 less than is the practice in many other places. The head under 
 which filtration takes place is seldom more than 0.5 meter (say 
 20 inches), but even with this low pressure and slow dehvery it 
 has been found impossible, with clean sand alone, to filter the 
 water satisfactorily. If the unfiltered water be allowed to stand 
 a fortnight or so, although the larger of the suspended particles 
 will have settled to the bottom, the water still retains a milky 
 appearance, and sand alone cannot remove the exceedingly fine 
 particles to which this appearance is due. On this account the 
 water from a! freshly cleaned filter is not used at once, but is al- 
 lowed to stand on the bed and then to pass through very slowly 
 until a thin coating has formed on the surface of the sand. This 
 coating is essential to the removal of the finest particles from the 
 water subsequently filtered. Of course, as the coating becomes 
 thicker the filtration becomes more difificult until it partially 
 stops and the filter is " dead." 
 
 As has been hinted above, the great trouble in summer is 
 from the abundance of small algae which soon clog the filter. 
 When the algse are absent, a square meter of surface usually fil- 
 ters 20 cubic meters before cleaning is necessary ; but in summer 
 the capacity is not over 10 cubic meters to the same area, and 
 when a slimy coating of decayed algae covers the surface of the 
 sand it becomes impossible, under the pressure commonly em- 
 ployed, to pass more than 2 cubic meters of water through i square 
 meter of sand surface. In the Berlin beds the thickness of the 
 sand, is 600 millimeters (about 2 feet). At each cleaning the sand 
 is removed for a depth of about 10 millimeters, but fresh sand 
 is not returned to the bed until only about 200 millimeters of the 
 original thickness remain. The foul sand is allowed to stand ex- 
 posed to the air until, by decay, the organic matter has lost its 
 slimy character ; it is then washed and eventually replaced upon 
 the beds. When the filters are emptied the water is drawn com- 
 pletely off, and by successive and systematic stirring nearly the 
 entire thickness of the sand is exposed to the air, in order that 
 the small amount of organic matter which was not retained at 
 the surface may be oxidized and destroyed. With the same 
 object in view, the air is allowed to circulate freely, for several 
 
PRACTICAL RESULTS OF FILTRATION. 
 
 1 59 
 
 days if possible, through the coarser underlying material. These 
 filters are filled from below with filtered water, and then the 
 water is passed through slowly and is allowed to waste for sev- 
 eral days. 
 
 Practical Results of Artificial Filtration. 
 As far as the suspended matter is concerned, the chief diffi- 
 culty in obtaining a bright and clear filtered water lies with the 
 finely divided clay which forms the turbidity of many streams, 
 especially at times of high' water. The Berlin experience has 
 already Been alluded ta. The following table, taken from the 
 Sixth Report of the Rivers Pollution Commission (p. 215), will 
 give an idea of the efficiency of the filtration as practised by the 
 various London companies. The observations being made on 
 monthly samples, the statements of the table will perhaps hardly 
 give a just' idea of the results obtained day by day ; but they 
 will serve to indicate the fact that the mere possession of filter- 
 beds does not secure perfectly clear water at all times. 
 
 TABLE XXII. — Thames and Lea Water — Comparative Efficiency of dif- 
 ferent Rates of Filtration during the years 1868 to 1873, inclusive. 
 
 
 Maximum rate 
 
 of Filtration 
 
 expressed in 
 
 inches per 
 
 hour. 
 
 Number of Monthly Occasions when — 
 
 Name of Compai5y. 
 
 Clear. 
 
 Slightly 
 Turbid. 
 
 T-bid. : ^Vcry^_ 
 
 Thames. 
 
 7.27 
 4.71 
 6.00 
 6.97 
 12.00 
 
 5.00 
 
 3-85 
 
 49 
 75 
 41 
 55 
 42 
 
 70 
 51 
 
 15 
 
 
 24 
 14 
 
 11 
 
 4 
 
 18 
 
 5 6 
 
 West Middlesex 
 
 
 
 5 
 
 7 
 
 12 
 
 
 
 Southwark and Vauxhall 
 
 Grand Junction 
 
 4 
 
 
 
 10 
 
 Lea. 
 
 ] 
 
 East London 
 
 3 2 
 
 
 
 
 As already explained, in addition to clarification of the wa- 
 ter, ordinary filtration does remove an appreciable amount of 
 organic matter previously held in solution. In addition to the 
 explanation given on page 153, it may be said that a part of the 
 effect which has been observed may be ascribed to the oxide of 
 iron,* and perhaps other minerals, which exist in the sand used 
 for filtration, the sand being seldom or never pure quartz. 
 
 The effect of filtration on th e water of the Thames and Lea 
 
 * This is the view of Thomas Spencer, the inventor of the so-called " carbide of 
 iron," used sometimes as a filtering medium . See page- 166. 
 
i6o 
 
 WATER SUPPLY. 
 
 has been made the subject of experiment by the Rivers Pollu- 
 tion Commission and others. The following table includes some 
 of the results obtained : 
 
 TABLE XXIIL — Observations on the Water of the various London 
 
 Companies. 
 
 [Results expressed in Parts per 100,000.] 
 
 Company. 
 
 New River. 
 New River, Stoke Newington 
 
 New River, New River Head 
 
 Thames River, 
 Southwark, Hampton 
 
 Chelsea 
 
 Lambeth 
 
 Grand Junction 
 
 Southwark, Battersea 
 
 West Middlesex 
 
 River Lea. 
 East London 
 
 Date. 
 
 < 
 1873. 
 
 Jan. 25. . 
 
 Jan. 
 
 27.. 
 
 Jan. 
 
 31.. 
 
 Jan. 
 
 3I-- 
 
 Jan. 
 
 3I-- 
 
 Feb. 
 
 3-- 
 
 Feb. 
 
 5.- 
 
 Feb. 
 
 7-- 
 
 Feb 
 
 I. . 
 
 3 So 
 
 few a 
 
 JUnf.. 
 ( Filt'd. 
 
 (Unf.. 
 ( Filt'd. 
 
 (Unf.. 
 I Filt'd. 
 
 (Unf.. 
 ( Filt'd. 
 
 (Unf.. 
 ( Filt'd. 
 
 (Unf.. 
 (Filt'd. 
 
 ( Unf.. 
 i Filt'd. 
 
 Unf.. 
 Filt'd. 
 
 Unf.. 
 Filt'd. 
 
 31.98 
 30.16 
 
 31.96 
 31.56 
 
 a 5 
 
 so 
 o 
 
 0.350 
 0.246 
 
 0.330 
 0.242 
 
 32.000.321 
 31.560.273 
 
 31.36 
 31.10 
 
 32.96 
 32.74 
 
 0.325 
 0.25S 
 
 0.273 
 0.258 
 
 31 .42 0.262 
 30.680.231 
 
 31-80 
 30.90 
 
 31-22 
 30.56 
 
 5P 
 O 
 
 0.084 
 0.042 
 
 0.061 
 0.043 
 
 0.063 
 0.042 
 
 0.046 
 0.032 
 
 0.067 
 0.038 
 
 0.042 
 0.032 
 
 0.2390.047 
 o.226'o.035 
 
 0.2090.071 
 o. 198 0.043 
 
 34.68 0.363 0.082 
 34 70' 0.305 0.041 
 
 i i 
 
 0.004 
 o 
 
 0.004 
 o 
 
 O.OOI 
 O 
 
 0.003 
 
 O 
 
 0.004 
 
 O.OOI 
 
 0.004 
 
 O.OOI 
 
 0.005 
 
 O.OOI 
 
 0.005 
 
 O.OOI 
 
 0.004 
 
 O.COI 
 
 From this table it appears that the filtration effects an appre- 
 ciable decrease in the amount of organic matter as judged from 
 the " organic carbon " and " organic nitrogen." In ordinary 
 practice this effect is trifling, and sand filtration is not sufficient 
 to remove the color which many surface waters possess, nor to 
 completely remove the unpleasant taste which sometimes affects 
 such waters. On these points the author has satisfied himself by 
 abundant experiments, and this is also the experience at Berlin. 
 Here, as has been mentioned, the Spree water often possesses to 
 
RESULTS OF FILTRATION. 
 
 l6l 
 
 a marked degree the brownish-yellow color common to streams 
 which flow through marshy or peaty regions, and to the water of 
 most impounding reservoirs. This color, with the taste which 
 at times accompanies it, gives rise to general complaint, but even 
 very slow filtration fails to remove it to any considerable extent. 
 The slight action which has been observed in this connection, 
 Piefke, resting on experiments made with prepared cellulose, 
 ascribes rather to the sediment containing vegetable fiber, than 
 to the sand itself. Table XXIV contains the results of a few 
 examinations of water from certain American localities : here 
 the amount of organic matter is indicated by the " albuminoid 
 ammonia," and this, when the filter beds are in good working order, 
 is appreciably less in the filtered than in the unfiltered water. 
 
 TABLE XXIV. — Examination of Water from Poughkeepsie and Hudson, 
 
 N. Y. 
 
 [Results expressed in Parts in loo.ooo.] 
 
 Q 
 > 
 
 U 
 td 
 K 
 
 Q 
 
 1877. 
 
 Nov. 13, 
 
 Nov. 13, 
 
 Nov. 19, 
 
 Nov. 19, 
 
 Nov. 27, 
 
 Nov. 27, 
 
 Dec. 10, 
 
 Dec. 10, 
 
 1878. 
 
 Jan. 18, 
 
 Jan. 18, 
 
 Locality. 
 
 POUGHKEEPSIE. 
 
 River 
 
 Clear-water basin 
 
 River 
 
 Clear-water basin 
 
 Hudson. 
 
 River 
 
 Filtered water. . . 
 
 River 
 
 Filtered water. . . 
 
 Top of filter bed, 
 
 i.e., unfiltered. . 
 
 Filtered water . . . . 
 
 
 < 
 
 Solid Residue. 
 
 K a: 
 
 < 
 
 
 s 
 s 
 
 < 
 
 s 
 
 
 
 ^2 
 
 
 
 is 
 
 ,0 ri <i 
 
 " Organic 
 
 and 
 Volatile." 
 
 Total at 
 212" 
 Fahr. 
 
 o.oiog 
 
 0.0197 
 
 1-7 
 
 12.1 
 
 10. I 
 
 0.0077 
 
 0.0139 
 
 I.I 
 
 9.1 
 
 9.0 
 
 0.0104 
 
 0.0157 
 
 1-5 
 
 10.5 
 
 8.6 
 
 0.0112 
 
 0.0155 
 
 1-3 
 
 9.4 
 
 g.o 
 
 0.0059 
 
 0.0152 
 
 I-I3 
 
 8.21 
 
 
 . 0040 
 
 0.0123 
 
 
 
 
 0.0051 
 
 0.0152 
 
 0.72 
 
 8.40 
 
 
 0.0056 
 
 0.0131 
 
 1.02 
 
 8.14 
 
 
 0.0123 
 
 0.0133 
 
 1.12 
 
 10.64 
 
 10.00 
 
 0.0237 
 
 0.0163 
 
 0.92 
 
 11.12 
 
 10.60 
 
 Remarks. 
 
 Very turbid. 
 Clear. 
 
 Very turbid. 
 Slightly turbid. 
 
 Turbid. 
 Slightly turbid. 
 
 Turbid. 
 Slightly turbid. 
 
 Turbid. 
 Slightly turbid. 
 
 Sand Filtration in the United States. 
 Up to the present time there has been very little done in this 
 country in the way of systematic filtration of water supplies, 
 partly, perhaps, from indiiTerence and lack of information, but 
 
1 62 WATER SUPPLY. 
 
 mainly on account of the expense. The numerous complaints 
 which arise in the case of almost every city and town supplied 
 with surface water render the question of filtration an important 
 one, and attempts have been made in various places to accom- 
 plish the desired object with a less expensive and elaborate plant 
 than that required by the English system. 
 
 Poughkeepsie, on the Hudson River, in the State of New 
 York, was the first city in the Union to adopt a scheme for 
 the artificial filtration of the entire water supply. The filtering 
 works consist * of a settling-basin 25 x 60 feet in plan and 12 feet 
 deep, in three compartments, arranged with reference to the 
 deposition of the heavier particles of mud before the water 
 passes on to the beds. The two filter beds are each 200 by 73^ 
 feet in plan, and 12 feet deep, built with vertical walls ; each has, 
 therefore, 14,700 square feet of filtering area. The 6 feet of 
 filtering materials, beginning at the top of the bed, are disposed 
 as follows : 
 
 24 inches of sand. 
 
 6 " " i inch gravel. 
 
 6 " " i " 
 
 6 " " t " 
 
 6 " "2 " broken stone. 
 24 " " 4 to 8 in. " " 
 
 Total, 72 inches. 
 
 The beds have a concrete bottom or floor 12 inches in thick- 
 ness, upon which are arranged open stone culverts to conduct the 
 filtered water to the intermediate basin. The flow of water from 
 each bed to. this intermediate basin is controlled by a gate, so 
 that while one bed is being cleaned the other may be used. The 
 filtration is conducted in the usual manner, as is also the clean- 
 ing and renewal of the sand, an inch or so of sand being removed 
 at a time, and the sand being washed and replaced only when 
 the upper layer has been much reduced in thickness. 
 
 The intermediate filtered-water basin is 6x85 feet in plan, 
 and 16 feet deep. This retains the filtered water until it is 
 allowed to pass into the filtered-water reservoir. This reservoir 
 is 28 X 88 feet in plan, and 17 feet deep, and from it the water is 
 pumped to the uncovered distributing reservoir from which the 
 
 * See Fourth Annual Report of the Water Commissioners of the City of Pough- 
 keepsie for the year ending Dec. 31, 1872. 
 
COVERED FILTER BEDS. 1 63 
 
 service pipes are fed. Sluice gates and drain pipes permit the 
 lowering of the water on the beds in any or all of the basins. 
 
 The city of Hudson, N. Y., is also supplied from the Hudson 
 River. The river vi^ater is pumped to the summit of a hill over- 
 looking the town, on which are situated the filter bed and the 
 distributing reservoir. The filter basin "'' is 13^ feet in depth, is 
 built with sloping sides, and has an area, at the surface of the 
 sand, of 9,081 feet. 
 
 The filtering material is six feet deep, and is arranged pre- 
 cisely as in the Poughkeepsie works which have been already 
 described. The fragments of broken stone rest upon a concrete 
 floor six inches in thickness, having a slight inclination toward 
 the middle or axial line, and this line toward the outlet. Along 
 this line runs an openly-laid stone culvert 18x24 inches, which is 
 connected by a cast-iron pipe under the division embankment 
 with the clear-water well. From the clear-water well the filtered 
 water passes over a gate or weir, where it is measured and its 
 flow regulated, to the clear-water basin or distributing reservoir. 
 Thence it passes ordinarily into the effluent chamber through 
 fine copper-wire screens to the 18-inch supply pipe ; but the 
 clear-water well can be connected directly with the supply main, 
 so that the city may be supplied from the bed without passing 
 the water through the basin or distributing reservoir. The dis- 
 tributing reservoir is 20 feet deep ; its capacity is 3,200,000 gal- 
 lons. Chemical examinations of the water from these localities 
 have been given in Table XXIV. 
 
 Advantages of Covered Filter Beds. 
 
 The exposure of a comparatively thin layer of water on the 
 surface of the filter beds has at least two disadvantages. In the 
 first place, in summer the water becomes heated and is, conse- 
 quently, in a condition to favor the growth of the lower orders 
 of plant life ; in the second place, in winter there is likely to be 
 inconvenience from the freezing of the water. In the climate of 
 England neither of these difficulties is as serious as in countries 
 which are either much warmer or much colder, and the filter 
 beds are universally uncovered. On the Continent, however, 
 
 * See Third Report of the Water Commissioners of the City of Hudson, 1875. 
 
164 "WATER SUPPLY. 
 
 the beds and the clear-water reservoirs are sometimes covered as 
 at Berlin, Magdeburg and other places. With reference to the 
 first point — vegetable growth — some trouble is experienced even 
 in England, and the beds become clogged with a confervoid 
 growth, which forms, as it were, a sort of carpet on the surface 
 of the sand, and this, when the beds are cleaned, can be raked 
 off or rolled up in a coherent sheet. This trouble might be 
 lessened somewhat by the use of covered beds, but where the 
 water to be filtered, contains an abundance of minute algae, as is 
 the case with the water of the Spree, at Berlin, there is no 
 perceptible difference in the condition of the covered and un- 
 covered beds. 
 
 With reference to the second point alluded to above — the 
 freezing of the water in winter — the European practice, in loca- 
 tions where the ice freezes to any thickness, may be learned by 
 the following quotation from Kirkwood's account of the Berlin 
 works : " The long and severe winters here made special care 
 and precaution necessary in the use of filters during the months 
 of severe frost. The filter beds cannot be laid bare in mid- 
 winter ; for the frost would in that case penetrate the body of 
 the filter and render it useless. All the filters are, in conse- 
 quence, during the winter months, kept constantly covered with 
 their maximum depth of water, four feet. Luckily the river 
 water during the winter months is in its best state as regards 
 freedom from turbidity, and also as regards freedom from vege- 
 table discoloration or impurity. The filters, therefore, have 
 comparatively little to intercept, and the river water is flowed 
 continuously upon them, and passes through them without 
 very sensibly impairing their efficiency. To make provision, 
 however, for an unusually long winter, or for an exceptional 
 condition of the river then, which may occasionally occur, it is 
 evident that a larger filtering surface is desirable than would be 
 necessary in a milder climate. 
 
 " The ice forms upon the filter beds 1 5 inches thick, and 
 sometimes, though rarely, 24 inches thick. To protect the en- 
 closing walls of each filter from damage, the ice is kept separated 
 from the walls, 6 to 12 inches, by attendants appointed to that 
 duty ; and, so long as the cake of ice is kept floating in this way, 
 the masonry is safe from any danger by its thrust. That this 
 service has been well performed, is demonstrated by the condi- 
 
FILTRATION IN WINTER. 1 65 
 
 tion of the walls, which are in the best of order, and nowhere 
 out of line, or abraded, that I could perceive." 
 
 Since the date of Mr. Kirkwood's report, covered filter 
 beds have been built, and it is stated that the uncovered beds 
 are not cleaned during the winter, the burden of the work 
 being thrown upon the covered beds. At Poughkeepsie and 
 at Hudson, the filtering area is not sufficient to deliver the 
 water throughout the winter without occasional cleaning. The 
 ice has therefore to be broken up and thrown, or rather, dragged 
 out. 
 
 There is no question but that water once filtered should be 
 distributed as soon as possible to the consumers. If it is neces- 
 sary that the water should be stored, it should be in covered 
 reservoirs of small size, which can be readily emptied and cleaned 
 in case of necessity. Apparently, the spores of certain algae are 
 not removed by filtration : at any rate, it has been found that if, 
 after filtration, the perfectly clear Spree water is allowed to 
 stand for eight or ten days, algse are developed. This fact is of 
 no practical consequence in a case like that of Berlin, where the 
 clear-water reservoir is too small to hold a single day's consump- 
 tion, and where, consequently, the water is delivered at once 
 into the service mains. 
 
 Expense of Sand Filtration. 
 
 The most valuable accessible data of the expense of filtration, 
 as drawn from actual experience, are found in the reports of the 
 Poughkeepsie Water Works. From these data it seems that the 
 ' expense may be set at from $2.50 to $3.50 per million gallons, 
 not allowing for the interest on the plant or for the cost of pump- 
 ing. The original cost of the beds was $54,000, the interest on 
 which would exceed the cost of maintenance. In 1879, Mr. J. 
 P. Davis, City Engineer of Boston, Mass., estimated the cost^of 
 constructing and operating artificial filters for the Mystic water 
 supply of the city — 10,000,000 gallons daily. He allowed for 
 seven beds, each with an area of 33,000 square feet, and esti- 
 mated that the cost of pumping and of operating the filters would 
 be about $5 per million gallons, and the interest on the neces- 
 sary works, at five per cent, would be nearly $6.00 per million 
 gallons, making the total cost about $11.00. 
 
1 66 WATER SUPPLY. 
 
 Filtering Materials other than Sand. 
 
 Many other substances have been proposed from time to 
 time as suitable to replace the sand wholly or in part, and to 
 accomplish more than sand can by chemical action on the impu- 
 rities of the water filtered. The so-called carbide of iron, of Mr. 
 Thomas Spencer, is used in several towns of England with some 
 success. The carbide of iron is prepared by roasting a mixture 
 of hematite iron ore and sawdust, and is held to consist mainly 
 of the magnetic oxide of iron : it is, no doubt, an efficient puri- 
 fying agent. It is, however, expensive, and could hardly be pre- 
 pared for less than $20 or $25 per ton, and for the best effect 
 should be preceded by a rough sand filtration. At Wakefield, 
 England, where, to be sure, the water is extremely filthy, and 
 the bed confessedly overworked, the Rivers Pollution Commis- 
 sion found that " the water, owing in part to putrescent fermen- 
 tation and subsidence, and in part to filtration, was chemically less 
 contaminated than might be expected, yet on both occasions it 
 contained a large proportion of nitrogenous organic matter. It 
 was of a greenish-yellow color, and on one occasion very turbid." 
 
 Various attempts have been made to use iron as a. filtering 
 medium since Medlock, in 1857, patented the process for purify- 
 ing water by allowing it to stand for some time in contact with a 
 considerable quantity of metallic iron. It is claimed that " spongy 
 iron " is now being used with success at Antwerp.* This mate- 
 rial, which was introduced as a medium for household filters a 
 few years ago by Prof. Bischof, is prepared by reducing hema- 
 tite ore, and is in a peculiarly porous or spongy condition. The 
 filters at Antwerp are said to have been laid out to treat over 
 two million gallons per day, but it does not appear that anything 
 like that amount is yet treated. The works went into operation 
 in June, 1881, and after twelve months Dr. Frankland was re- 
 quested to examine and report on them. The following is taken 
 from his report : 
 
 " The water, which was abstracted from the river Nethe, about 
 fifteen miles above Antwerp, is first impounded in two reservoirs, 
 where it is allowed to subside for from 12 to 24 hours ; from these 
 reservoirs it is pumped on to the spongy iron filters, whence it 
 flows by gravitation upon sand filters. 
 
 * Circular of " The Spongy Iron Water and Sewage Purifying Company," London. 
 
FILTRATION THROUGH SPONGY IRON. 167 
 
 " The spongy iron filters consist of two layers of bricks loosely 
 laid upon a bed of concrete. On the bricks rests a layer 3 feet 
 in thickness, formed of 5 m.m. gravel mixed with one-third of 
 its bulk of spongy iron. Then comes a layer 3 inches thick of 
 fine gravel, and lastly a stratum of sand 2 feet deep, making in 
 all 5 feet 3 inches of filtering material. 
 
 " The sand filters are similarly laid upon bricks and concrete. 
 They consist of a layer of 5 m.m. gravel i foot thick, covered 
 with 3 inches of fine gravel and topped with 2 feet 6 inches of 
 sand, making altogether 3 feet 9 inches of filtering material. 
 
 " The area of filtering surface of each filter amounts to 7,302 
 square feet, and the rate of filtration varies from 300 to 500 
 gallons per minute, or from 60 to 100 (imp.) gallons per square 
 foot per 24 hours. 
 
 " The result of the analysis of the three samples of water 
 show that even after subsidence for nearly 24 hours, the water 
 of the Nethe is exceedingly impure, being still turbid and loaded 
 with an unusually large proportion of highly nitrogenized organic 
 matter. The composition of the water as it passed on to the 
 spongy iron filters is stated to have been: 
 
 Total solids (mostly dissolved) 21 parts in 100,00a. 
 
 Organic Carbon o. G23 
 
 Organic Nitrogen o,2ig 
 
 Ammonia o 028 
 
 Chlorine (combined) 1 ■ 8 
 
 Hardness — temporary 4-^° 
 
 ' ' permanent 6-9° 
 
 total 11-5° 
 
 The water in this condition was very unpalatable. 
 
 " The aggregate effect produced by one filtration through 
 spongy iron was as follows : 
 
 Total percentage reduction. 
 
 Total solids 4i-3 
 
 Organic carbon 60. 9 
 
 Organic nitrogen 74-9 
 
 Ammonia 
 
 Total combined nitrogen 77-3 
 
 Chlorine > °' 
 
 Temporary hardness i3-0 
 
 Permanent " 35-3 
 
 Total " 27.0 
 
1 68 WATER SUPPLY. 
 
 " The nitrogenous character of the organic matter was dimin- 
 ished from the initial proportion, nitrogen to carbon = i : 2.84, 
 down to I : 4.4. By boiling, the hardness of the doubly filtered 
 water is reduced to 4.4 parts per ioo,ooo, or 3° on Clark's scale. 
 
 " Lastly, from being muddy, unpalatable, colored and much 
 polluted, the water of the Nethe was rendered colorless, bright, 
 palatable and fit for dietetic and domestic purposes." 
 
 Wood-charcoal is often used in filters of small size, mixed 
 with sand. Practically, however, it adds nothing to the efficiency 
 of a properly managed sand filter. One way in which charcoal 
 is used is illustrated by the works of Marshalltown, Iowa. Here 
 a filter basin, 32 x 16 feet, was built of masonry ; and a filter floor 
 of two-inch plank was supported on joists laid crosswise. The 
 floor was pierced with three-fourth inch holes, and covered with 
 wire gauze. On this there is a layer of charcoal four inches thick, 
 and above this 14 inches of clean gravel and sand. At Clinton, 
 Iowa, a number of boxes, 16 in fact, filled with charcoal, gravel, 
 and sharp sand, rest upon the conduit. The water flows on to 
 the boxes, and through the material into the conduit. The boxes 
 can be raised one. at a time for cleaning. In case of fire, how- 
 ever, the water is taken into the conduit without filtration. 
 Sponge, which is much used in filtering water for manufacturing 
 operations, such as paper-making, has been used to a limited ex- 
 tent in connection with sand and gravel, even on the larger scale 
 of a town supply. Alton, 111., pumps from the Mississippi River; 
 and the water is filtered through sponge contained in a cast-iron 
 filter box of 54 cubic feet capacity : this box fits into a tight 
 chamber in the aqueduct leading from the river to the pump- 
 well, and can be raised by machinery. The box can be raised, 
 the sponges renewed, and the box replaced, in three hours. The . 
 amount of water filtered is about 150,000 gallons a day. When 
 the river is muddy the sponges are cleaned every three or four 
 weeks : sometimes, when the river is clear, not oftener than once 
 in three months. An attempt was made at one time to filter the 
 supply of the village of Malone, N. Y., through a filter of soft 
 brick, but it was not found practicable to filter with sufficient 
 rapidity. 
 
 Many other water works, in this country, make some attempt 
 to " filter " their water by passing it through broken stone, gravel, 
 or gravel and charcoal, or even through sand and gravel. In 
 
HOUSEHOLD FILTRATION. 169 
 
 general, the most that can be said of these arrangements is that 
 they act with greater or less efficiency as strainers, removing 
 some of the coarser matters ; the infrequency of the cleansing 
 showing that the work done cannot be very great. As an illus- 
 tration of this point, may be mentioned a locality where the filter 
 beds were constructed as long ago as 1853. They are built with 
 sloping sides and measure 50x60 feet. The filtering material, 
 which consists of sand, gravel, and pebble stones, has an entire 
 thickness of 24 inches, and filtration is carried on under a head 
 of from 10 to 15 feet. The beds are not used in winter, but when 
 in use the amount of water filtered daily is 1,500,000 gallons. 
 The beds are cleaned not oftener than once a year. This is an 
 extreme case, but inadequate area and infrequent cleansing are 
 the common faults of many so-called filters. Of course, occa- 
 sionally, the character of the suspended matter which is to be re- 
 moved is such that a very simple straining process is all that is 
 required : this is the case with some of our streams on which are 
 a number of saw-mills, and where the comparatively coarse par- 
 ticles of sawdust comprise the main part of the impurity. In 
 such a case, as, for instance, at Eangor, Me., simple passage 
 through a limited amount of sand is all sufficient. 
 
 Household Filtration. 
 
 In localities where there is a public water supply, it is, with- 
 out doubt, the duty of the water board or company to deliver 
 the water to consumers in a condition fit for domestic use. If 
 the source which is, on the whole, the most available for the 
 water supply is such that filtration is absolutely necessary, the 
 water should be filtered on the large scale by the authority con- 
 troUing the works. Practically, howeyer, in the case of most 
 existing water supplies, the water as delivered to the consumers 
 may be appreciably improved by filtration ; household filtration 
 is also often necessary in country residences and in the smaller 
 towns where there is no public supply, and where it is necessary 
 to use rain water which has been stored in tanks or cisterns. 
 
 For filtration on the household scale, numerous devices have 
 been made and patented, and. the greatest variety, of material 
 has been proposed : many sorts of porous stone, sand, powdered 
 glass, bricks, iron in turnings and other forms, vegetable and 
 animal charcoal, sponge, wool, flannel, cotton, straw, sawdust, 
 
I/O WATER SUPPLY. 
 
 excelsior and wire-gauze — these are some of the substances 
 which are used. A filter suitable for household use must be 
 made of a material which cannot communicate any injurious or 
 offensive quality to the water which passes through it ; it must 
 remove from the water all suspended particles, so as to render 
 the water bright and clear; and it must either be readily cleaned, 
 or the filtering material must be such as to be readily renewed. 
 In addition to these requirements, it is of great advantage if the 
 filter is able to remove a noticeable amount of the dissolved 
 organic matter which most waters contain. 
 
 As to the filtering material, the author* is satisfied that 
 there is nothing, on the whole, better than well-burned animal 
 charcoal (bone-coal). This material, as is well known, possesses 
 great power in removing organic matter from solution, and is 
 used in the arts to decolorize colored solutions : on rtiany organic 
 substances it acts, not simply by adhesion, but apparently by 
 bringing them into contact with oxygen, and thus absolutely 
 destroying them. Its power does not last indefinitely, and a 
 bone-coal filter, like a filter of any other material, requires cleans- 
 ing and renewal at proper intervals. Other materials to be 
 mentioned render good service, and in certain sorts of filters, 
 as, for instance, those made for attachment to ordinary cocks 
 or faucets, the bone-coal possesses no essential advantage. 
 
 We will consider first the filters of small size, intended to 
 be attached to the faucet, where the water is brought in pipes 
 either from the service-mains of a general supply, or from a 
 tank in the building ; second, the portable filters intended to 
 occupy a more or less permanent position, and to,be filled with 
 water, either by some ball-cock or other similar arrangement, 
 or by means of smaller supplies continually renewed ; third, 
 the more permanent and fixed devices which are inserted or 
 built into underground and other cisterns, or are introduced 
 into the course of the service-pipes so that all the water used 
 in the house passes through them. 
 
 Considering the volume of water which must flow through 
 an extremely limited amount of material, no filter capable of 
 being screwed on to an ordinary water-tap can act in any other 
 way than as a strainer, and all that can be required of such a 
 filter is that it shall remove suspended particles and be readily 
 
 * See Nichols : Filtration of Potable Water. New York, Van Nostrand. 
 
HOUSEHOLD FILTRATION. 
 
 171 
 
 cleansed or renewed at trifling expense. The older forms of 
 filters, which could be cleaned only by unscrewing from the 
 tap and reversing either the whole apparatus or some inside 
 receptacle of the filtering medium, were all open to objection, 
 and no one of them was to be recommended as superior to 
 the primitive and unpatentable device of attaching to the faucet 
 a bag of cotton- 
 flannel to be fre- 
 quently washed 
 and renewed. At 
 present, however, 
 there are certain 
 forms of filter in 
 the market which 
 can be reversed 
 without removing 
 them from the tap, 
 and which are su- 
 perior to anything 
 previously in use 
 for the purpose. 
 Figure 32 shows 
 such a filter. The 
 water passes 
 through the wire 
 gauze at E and 
 the filtering material F. By means of a handle, the stem of which 
 is shown in the figure, the ball may be turned over and the sedi- 
 ment collected on E be carried away with the first rush of water. 
 By turning the handle half way, water may be drawn directly 
 without passing through the filtering-ball. 
 
 We come now to the larger forms of filters, to those which 
 are portable, but which are intended to occupy a permanent 
 position in the room, or in some cases to be placed in the tank 
 from which the supply is drawn. The material which, next to 
 simple sand, has probably been used as long as anything for the 
 purpose, is stone. Some varieties of sandstone are particularly 
 porous, sufficiently so to allow of the use of slabs of the stone 
 as filters ; other similar substances, such as pumice stone or 
 unglazed earthenware, have been employed ; the most common 
 
 Fig. 32. 
 
1/2 
 
 WATER SUPPLY. 
 
 arrangement being to insert the stone as a horizontal partition 
 in a small tank or vessel : the action is, in the main, mechanical, 
 and the sediment which collects upon the upper surface of the 
 block is removed by washing and brushing. A material which 
 is used to a considerable extent in England is the so-called sili- 
 cated carbon : it is the residue of the distillation of a certain 
 variety of bituminous shale. Thus it is a coke mixed with min- 
 eral matter, and is compressed into blocks for use. In the com- 
 mon form of household filters of this make, the block is cemented 
 as a partition into an earthen jar, and is not readily cleaned, biit 
 there is no doubt that, until the filter becomes clogged, it is 
 very efficient in purifying water even from the dissolved organic 
 
 matter. 
 
 As a type of the better 
 class of tank filters, we may 
 take that much employed in 
 England and manufactured 
 by the London and General 
 Water Purifying Company; 
 this filter* is shown in ele- 
 vation and in section in Figs. 
 33 and 34. The earthenware 
 filter box is filled with ani- 
 mal charcoal, in the form of 
 charred bones, broken into small pieces, and freed from dust. 
 There is no chamber for storing filtered water : the water is 
 filtered at the time it is drawn off for use. The filter is readily 
 cleansed and the charcoal re- 
 newed. The water passes 
 upward through the filtering 
 material, in order that mat- 
 ters spontaneously settling 
 down may not be deposited 
 upon the filtering material, 
 and may not, therefore, help 
 to clog its pores ; and, fur- 
 ther, that the suspended 
 matters strained from the 
 water, being separated as they 
 
 * A very similar filter made in Boston, England, is on salein this country. 
 
 Fig. 33. 
 
 Fig. 34. 
 
HOUSEHOLD FILTRATION. 
 
 173 
 
 always are, mainly at the surface of the filtering material, may 
 fall away from it and deposit elsewhere ; the consequence of this 
 is, that the filtering material requires less frequent cleansing. 
 
 Bone coal is also used compressed into cylindrical blocks 
 with an aperture in the center into 
 which the delivery pipe is inserted, 
 and one or more of these may be 
 placed in the tank from which the 
 water is taken. As a rule, the pre- 
 viously described method of using 
 the bone coal is to be preferred. 
 
 Another material which has late- 
 ly come into use to a considerable 
 extent in England, is what is known 
 as " spongy iron," which has already 
 been alluded to in the description 
 of the filtration works at Antwerp. 
 The portable filters are constructed 
 in various forms, but on the same 
 general plan. Fig. 35 represents 
 one form, where the water is sup- 
 plied from an inverted bottle, which 
 must be refilled as often as empty. 
 In other forms the reservoir of un- 
 filtered water is kept full by being 
 connected with the service pipe by 
 means of a ball-cock attachment. 
 The vessels are of earthenware, but 
 the spongy iron is also furnished in 
 filters which can be inserted in tanks 
 or cisterns. 
 
 Although, at any rate with the 
 smaller forms of filter, it is difficult 
 or impossible to obtain a water free from iron, there is no doubt 
 that a considerable portion of the dissolved organic matter is 
 removed, and it is claimed that bacteria and bacterial germs are 
 completely removed. These claims are borne out by the experi- 
 mental investigations of Bischof,* Hatton,t and others, and give 
 
 * Proc. Roy. Soc, xxvii, p. 258. 
 f Journ. Chem. Soc, xxxix, p. 247. 
 
174 WATER SUPPLY. 
 
 to the material a great theoretical advantage. Practically, how- 
 ever, it is extremely doubtful whether in the ordinary use of the 
 spongy iron in filters such results can be obtained ; and at the 
 best, the water must pass very slowly through the filter in order 
 that purification may take place. Like other materials used for 
 the purpose, it affords a means for improving an undesirable 
 water, and for lessening the risk in using a doubtful water : it 
 does not afford an excuse for employing a water known to be 
 impure, on the ground that possible danger will thus be certainly 
 averted. 
 
 Wood-charcoal is sometimes used in household filters, but, 
 practically, it has next to no chemical action. The author had 
 occasion some time since to examine an American filter which is 
 much used in certain sections of the country, and in which the 
 filtering medium is wood-charcoal with clean quartz pebbles. 
 The filtering material is arranged in an oak tub and the water is 
 placed in a zinc pan at the top, and passes first through a hand- 
 ful of sponge and then downward through the pebbles and char- 
 coal. The experiments extended over several months, and the 
 water was examined by Frankland's and by Wanklyn's method. 
 In the case of the Boston water, when the filter- was in constant 
 use, absolutely no effect was produced on the water — the water 
 which issued from the filter containing exactly the same amount 
 of organic matter as when it entered. If the sponge were taken 
 out, thoroughly washed and replaced, there was for a short time 
 a slight difference between the filtered and unfiltered water, but 
 the effect was temporary and soon ceased. In spite of the bulk 
 and weight of this filter (the smallest size of which weighs about 
 1 50 lbs.), the entire work with water of this character seems to 
 be done by the handful of sponge at the top. The filter is one 
 of a class which claim to be chemical in their action, but which 
 cannot do more than remove suspended matter ; and the action 
 on clayey waters, with which the filter is sometimes used to ad- 
 vantage, is probably largely due to the principle illustrated on 
 page 153 (Fig. 30). 
 
 The following simple form of filter is described by Dr. Smart, 
 and answers fully as well as many patented contrivances. " The 
 filter is made of tin and consists of a modified funnel, the body 
 of which rests on a tin bucket or receiver, while the tube projects 
 downward to the bottom of the said bucket. The lower end of 
 
HOUSEHOLD FILTRATION. 1 75 
 
 ♦ 
 
 the tube is tied over with some filtering cloth. Three-fourths of 
 its length is filled with granulated bone charcoal and the upper 
 fourth with sand. The upper end of the tube projects about half 
 an inch into the body of the funnel to permit of tying a filtering 
 cloth over the top of the sand. The angle between this projec- 
 tion and the sloping sides of the funnel will serve to trap solid 
 matter. To clean this filter, the filtering cloth guarding the top 
 of the tube will have to be removed, washed, and replaced. At 
 longer intervals, when the filter shows signs of clogging, half an 
 inch of the upper layer of sand may be removed and replaced by 
 fresh material. At yet longer periods, depending upon the length 
 of time during which the charcoal retains its powers of oxidation, 
 the whole contents of the tube may be dumped out and re- 
 newed. Earthenware is more durable than tin, and would pre- 
 serve the water cooler during the warm months." 
 
 There are various filters in the market which are arranged for 
 use where there is a public water supply, by being connected with 
 the service pipes in such a way that all the water entering the 
 building passes through the filter. By proper arrangement of 
 valves it is possible to reverse the current and cleanse the filter 
 from the collected impurities. In some cases the valves are so 
 arranged that the filter can be cleaned by a reverse current of 
 hot water or steam. The revolving filters, mentioned on page 
 171 as adapted to faucets, are also made on the larger scale for 
 insertion in the service-pipes of houses and other buildings and 
 for manufacturers' use. 
 
 We come now to the discussion of filters suitable for cisterns 
 of considerable size, and especially for the underground cisterns 
 in which rain water is usually stored. The collection of water 
 from the roofs of houses involves the collection of dust and dirt 
 more or less objectionable in character, especially in places where 
 soft coal is burned. Although it is possible by automatic con- 
 trivances to avoid the collection of the first portions of the water 
 coming at any time from the roof, yet these do not perfectly ac- 
 complish their intended object, and are not at all commonly em- 
 ployed. Moreover, the construction of ordinary cisterns is such 
 that, after the water is once collected, it is liable to deterioration 
 and to contamination by various foreign matters which fall into 
 it, so that, if not absolutely necessary, filtration is certainly very 
 desirable. Where the water is stored in tanks in the roof of the 
 
1/6 
 
 WATER SUPPLY. 
 
 building, one of the various-forms of filters just alluded to may 
 be placed beneath the tank and so connected that all the water 
 used shall pass through it. The outlet pipe from the tank should 
 start several inches from the bottom, in order that the sediment 
 may deposit itself as far as possible on the floor of the tank and 
 not be drawn into the filter: the tank should, of course, be 
 cleaned from time to time. The disadvantage of this method is 
 the inability to command a sufficient head of water to properly 
 clean the filter by reversing the current. The tank may be 
 divided by a partition and the water be required to pass through 
 a filter constructed in the tank itself and filled with sand and 
 charcoal or bone-coal, or one of the patent tank filters already 
 described may be employed. (See Figs. 33 and 34.) 
 
 With underground cisterns, it is not uncommon to construct 
 
 them so that the water is not 
 pumped directly from the cis- 
 tern, but from a sort of pump- 
 well, to enter which the water 
 must pass through a porous 
 partition wall made of bricks. 
 These walls are constructed in 
 various ways : one form is rep- 
 resented in the accompanying 
 cut, taken from " Scribner's 
 Monthly Magazine" for Sep. 
 tember, 1877. 
 
 When the brick partition is 
 new, it is undoubtedly of good service ; but it soon becomes 
 clogged, and covered on the outside with a deposit of organic 
 matter, so that after a time the water which passes through the 
 brick wall must first have an opportunity to leach out what it 
 can from this mass of decaying matter.* As a rule, the interiors 
 of cisterns are not very accessible, and when the cistern is relied 
 upon as the sole or as a principal supply for the household, it is 
 impossible to renew frequently the filtering wall, or even to 
 thoroughly clean the outer surface. The best that can be done 
 under ordinary circumstances is to clean the outer surface of the 
 wall as thoroughly as may be with a stiff brush every few months, 
 
 * Some analyses of cistern waters thus filtered have been given in Table V, page 51. 
 
HOUSEHOLD FILTRATION. 
 
 177 
 
 and to renew the wall completely whenever the probabihty of a 
 rainy season allows. If the body of the cistern be divided by a 
 partition wall into two compartments which may be made to 
 communicate or not at v/ill, the two may be cleaned at different 
 times and thus the danger of a water-famine be averted. 
 
 Other methods for accomplishing filtration in the cistern 
 
 have been proposed. Filters 
 of sand and charcoal, or of 
 bone-coal, are sometimes con- 
 structed within the cisterns; 
 but they are not easily reach- 
 ed for cleaning, and as a 
 rule they are allowed to go 
 uncleaned. Various devices 
 
 Fig. 37 (from Fischer). Fig. 38. 
 
 have been suggested for attachment to the suction pipe of the 
 pump, two of which are shown in Figs. 37 and 38. Fig. 37 is a 
 German device for using bone-coal compressed into blocks : Fig. 
 38 is an American device for taking the water as free from the 
 sediment as possible. The cylinder contains silicious sand for 
 the filtering medium and is buoyed up by an air-tight chamber 
 at one end. The pipe has a swivel-joint which allows the filter 
 
178 
 
 WATER SUPPLY. 
 
 to accommodate itself to the level of the water in the cistern or 
 reservoir.'"" 
 
 Still another method of accomplishing the desired object con- 
 sists in placing the filtering material in a frame capable of sliding 
 in a groove and of being readily lifted from its place. The fil- 
 tering material may consist of porous tiles or of blocks of animal 
 charcoal ; and, if duplicate frames are provided, the grooves may 
 be so arranged that a fresh frame can be lowered into place be- 
 fore the old one is taken away. Figure 39! represents a cistern 
 constructed with such frames containing blocks of animal char- 
 coal, as prepared by Atkins & Co., London. These blocks can 
 be readily^cleaned by scraping the outer surface (at some expense, 
 to be sure, of the material of the blocks), and they can be 
 renewed when necessary. They are made of various densities; 
 the most dense permitting the passage of 30 to 40 gallons per 
 square foot per day, while the most porous pass some 250 to 300 
 
 Fig. 39. 
 
 gallons. For use in ordinary cisterns tolerably porous blocks 
 would probably answer well enough, and for such use as this the 
 
 * Scientific American, Jan. 10, 1880. 
 
 ■f This cut is taken from Fanning's Water-supply Engineering. 
 
FILTRATION FOR MANUFACTURING PURPOSES. 1 79 
 
 charcoal is more conveniently employed in this form of blocks 
 than as fragments. 
 
 The arrangement which has been described is rather expen- 
 sive for common use ; although, if the necessary provision were 
 made in the original plan for the construction of the cistern, it 
 would, on the whole, be more satisfactory than other plans which 
 involve less outlay at the start. The author is not aware that 
 the blocks of compressed animal charcoal are prepared in this 
 country, but there would probably be no difificulty in obtaining 
 them if there were any demand. 
 
 There is one point worth noting in connection with domestic 
 filtration, namely, that in this country we are in the habit of 
 putting ice directly into the pitchers or small tanks from which 
 drinking water is served. Natural ice is not always clean, and 
 frequently, after the ice is melted, the water, even if clear at 
 the start, will be found full of suspended particles, or having an 
 abundant sediment. Filters are made, so that the ice cools the 
 water before the filtration takes place and the difficulty can also 
 be obviated to a large extent by inclosing the ice in a clean flan- 
 nel or cotton-flannel bag. 
 
 Filtration for Manufacturing Purposes. 
 
 For the majority of manufacturing purposes the object of 
 filtration is accomplished if a thorough removal of suspended 
 matter is effected. Sand, wood-charcoal, bone-coal, flannel and 
 various other substances are employed as filtering media ; very 
 good results are obtained by the use of sponge, and this mate- 
 rial is employed to a considerable extent ; in some cases, the 
 water is filtered through sand filters, similar to those used in 
 connection with town supplies but on a smaller scale. 
 
 Two filters have recently been offered in the market to effect 
 rapid filtration for manufacturing purposes and also for town 
 supply. The " Multifold Filter," manufactured by the Newark 
 Filter Company, may be of various sizes, and each apparatus 
 consists of a number of shallow cast-iron pans -ranged one above 
 the other. Each pan has a false perforated bottom, on which 
 rests a layer of sand, 6 inches or less in thickness, and each 
 works as an independent filter. The water passes in the di- 
 rection of the arrows shown in the cut under a greater or less 
 pressure. The chief peculiarity of the filter consists in the ar- 
 
i8o 
 
 WATER SUPPLY. 
 
 rangements for cleansing. When the sand becomes clogged, 
 
 water is introduced 
 under pressure 
 through the hollow- 
 axis of the cylinder, 
 and issues in fine 
 jets from the radial 
 -^ arms, stirring up the 
 sand and washing 
 away the lighter sub- 
 stances. The arms 
 can be revolved from 
 the outside, so that 
 the sand may be com- 
 pletely washed. 
 
 Another filter 
 (system Piefke) has 
 ^^'^- 40. been recently adver- 
 
 tised in Germany,* where the filtering medium consists of cellu- 
 lose (vegetable fiber) which has been impregnated with some anti- 
 septic substance. This material is disposed on a number of cir- 
 cular sieves arranged one above another in an apparatus some- 
 what similar to the preceding, and a thin layer suffices. The 
 filter is cleaned by revolving the vertical axis of the apparatus, 
 which is a rod to which scrapers are attached ; the water contin- 
 ues to flow through the filters, and the material is kept in sus- 
 pension until the impurities are washed away. This involves 
 some loss of the prepared cellulose, but it is claimed that the 
 filtration is efficiently and cheaply accompHshed. No details from 
 disinterested sources are at hand with reference to the practical 
 value of either of these devices, which seem to possess certain 
 merits. 
 
 * Journal fUr Gasbeleuchtung und Wasserversorgung, March, 1883. The appa- 
 ratus is figured and described in the Scientific American, April 28. rSSs. 
 
CHAPTER IX. 
 
 ARTIFICIAL IMPROVEMENT OF NATURAL WATER {Continued). 
 The Softening of Hard Water. 
 
 As has already been explained (page 33), the hardness of 
 water is generally due to the presence of compounds of lime or 
 magnesia. While a moderately hard water may be perfectly 
 well suited for drinking, for almost all the other purposes of a 
 water supply a soft water is preferable, other things being equal. 
 If common soap be added to hard water the water seems to 
 curdle, but no permanent froth or lather is formed until, by the 
 mutual action of the soap and the compounds of lime (and mag- 
 nesia) on each other, the latter are completely converted into a 
 lime (or magnesia) soap, an insoluble substance which forms the 
 curd alluded to. After this point is reached, any additional soap 
 becomes available for washing, but the curdy water is less effi- 
 cient as a detergent. Hard water is, as a rule, much less desir- 
 able for culinary purposes than soft water. Finally, hard water 
 is also objectionable on account of the " scale " which forms in 
 steam boilers in which it is used : in manufacturing towns this 
 becomes a matter of great importance. 
 
 Temporary hardness. — The temporary hardness is due to the 
 presence of carbonate of lime or magnesia : these compounds 
 are soluble in water to a slight extent only, but are brought into 
 solution by carbonic acid, as has been explained on page 9 : it 
 will be convenient to consider them as existing in solution as 
 ^z'carbonates. The temporary hardness of a water may be re- 
 moved in various ways : in the first place, by adding soap, as is 
 actually done when an attempt is made to wash with hard water 
 — a method uneconomical on the small scale, and impracticable 
 on the large scale on account of the expense ; in the second 
 place, by adding ordinary washing-soda (carbonate of soda) — a 
 method employed very generally on the small scale, but also 
 impracticable on the large scale on account of expense. The 
 
1 82 
 
 WATER SUPPLY. 
 
 chemical explanation of the second method is this : when car- 
 bonate of soda is added to water containing bicarbonate of lime 
 there result bicarbonate of soda and carbonate of lime ; the former, 
 is soluble in water and remains in solution, the latter being in- 
 soluble separates as a fine powder. A simpler method still, as 
 the explanation of the term " temporary " shows, consists in 
 boiling the water for half an hour or more : the (J2carbonate of 
 lime (or magnesia) is decomposed, losing half its carbonic acid ; 
 this carbonic acid escapes as gas, and the simple carbonate of 
 lime separates as a white powder. On account of this action, 
 carbonate of lime is one of the chief constituents of boiler scale, 
 and a similar deposit forms in the water-backs of kitchen ranges, 
 and, in fact, in any vessel in which the water is boiled. Some 
 natural waters are so highly charged with carbonate of lime that 
 slight agitation suffices to drive off the " extra " carbonic acid 
 and to allow the carbonate to separate ; large deposits of car- 
 bonate of lime occur in nature which owe their origin to such an 
 action. Fig. 41 (from the Scientific American) shows a portion 
 of the feed-pipe of a boiler which was nearly choked up by the 
 calcareous matter. 
 
 From a water which possesses temporary hardness, the car- 
 bonate of lime may 
 be caused to deposit, 
 and the water thus 
 become softened, not 
 simply by expelling 
 the " extra " carbonic 
 acid by heat or other- 
 wise, but also by caus- 
 ing this carbonic acid 
 to unite chemically 
 with some substance capable of thus decomposing the bicarbon- 
 ate. Caustic soda or caustic potash will produce this effect, but 
 the cheapest and most available substance is ordinary lime. The 
 lime unites with the extra carbonic acid to form carbonate of 
 lime, which settles out as a fine powder along with the carbonate 
 originally held in solution. As carbonate of lime is, not abso- 
 lutely insoluble in water, a small amount remains in solution 
 after the softening has been completed, not enough, however, to 
 be seriously objectionable. 
 
SOFTENING OF HARD WATER. 1 83 
 
 ; The use of lime was invented and patented about the year 
 1844 by Thomas Clark, professor of chemistry in the University 
 of Aberdeen, but the patent expired long since. The process 
 has been used in England at works furnishing as much as 
 1,000,000 gallons daily. The proper amount of lime is added in 
 the form of lime-water or milk of Hme. After thorough mixing 
 the water is allowed to subside for from 12 to 24 hours, and 
 drawn off from the sediment. The readiness with which the 
 finely divided carbonate of lime settles depends somewhat upon 
 the character of the water ; and, as it settles, it drags down 
 with it and removes from the water a not inconsiderable pro- 
 portion of the organic matter present ; if the water is colored 
 by peaty matter, a very appreciable decolorization is effected. 
 Experience, however, would seem to show that the process gives 
 the best results with water which is naturally clear, such as 
 spring water ; and, in the case of turbid river waters, the soft- 
 ening process should be followed by filtration. 
 
 The economy of the process and the advantage of softening 
 a hard water on the large scale, rather than by the use of soap in 
 the household, is evident when we consider that, to soften a 
 quantity of water requiring one hundred-weight of quicklime, the 
 expense of materials would be (approximately) : 
 
 I cwt. of lime, say $ u • 5° 
 
 5 cwt. of sal soda (at 1.2 cents per lb.) 6-00 
 
 20 cwt. of soap (at 6^ cents per lb.) 130.00 
 
 Of course, on a large scale the cost of labor, and especially of 
 handling the sludge, may make the actual difference less than the 
 theoretical, but, in any event, the saving in soap by the use of 
 the softened water is very great. 
 
 On account of the difficuhy with which the carbonate of lime 
 settles in waters containing much organic matter, of the length 
 of time required and of the necessarily large area of settling- 
 basins, the process is seldom carried out according to the original 
 plan. There are various modifications of the process which aim 
 to accomplish the object with greater economy of time and space, 
 using, however, the same material. These modified processes 
 involve some form of filter by which the precipitated carbonate 
 of lime may be removed at once without waiting for the sub. 
 
1 84 
 
 WATER SUPPLY. 
 
 sidence to take place. It will suffice to describe one of these 
 processes. 
 
 In England a number of manufacturing establishments and 
 several towns have introduced Clark's process as modified by 
 J. H. Porter, C. E.* The lime is employed in the form of a 
 saturated solution, and the mixing with the bulk of the water to 
 be softened takes place in a separate tank from that in which the 
 solution is prepared. When thorough mixture has been effected, 
 the liquid is at once filtered through cloth. The arrangement 
 and construction of the tanks varies with the quantity of water 
 
 Fig. 4j. 
 
 to be softened. Fig. 42 represents an apparatus for treating 300 
 gallons of water per hour. The preparation of the lime-water 
 
 * The PortT-CIark Process for the Softening, Purification and Filtration of Hard 
 Waters, by John Henderson Porter, London. 
 
SOFTENING OF HARD WATER. 
 
 185 
 
 takes place in the left-hand cylinder, which is furnished with a 
 mechanical stirrer ; the mixing takes place in the right-hand 
 cylinder ; the other details of the apparatus are evident from 
 the figure. The filtration of the mixture may be accomplished 
 by any one of a variety of filter presses ; that employed by Mr. 
 Porter is shown in Fig. 42 and more plainly in Figures 43, 44 
 and 45. Fig. 43 shows a portion of the filters used by the Lon- 
 
 FiG. 43. 
 
 don and North-western Railway Company at Liverpool, where 
 over 200,000 U. S. gallons of water are softened daily. Each 
 filter is made up of a series of cast-iron plates and cast-iron open 
 frames of the form shown in Figs. 44 and 45. " Over these filtering 
 
 Fig. 44. 
 
 Fig. 45. 
 
 chambers, of about i inch in thickness, is dropped (as a towel plac- 
 ed upon a towel-horse) a cloth of superior quality of cotton twill, 
 having worked in it holes to correspond with the holes through the 
 upper corners of both water-space frame and filtering chambers. 
 When these alternate water spaces and filtering chambers with 
 
1 86 WATER SUPPLY. 
 
 the cloths are tightly pressed together by a powerful end screw, 
 it will be seen that the holes become, collectively, tubular chan- 
 nels of the length of the ' battery,' the channel of the one side 
 admitting its chalky water to the circular water-spaces, whence, 
 being inclosed and under pressure, it can only escape through 
 the adjoining cloths into the concentric and radiating grooves 
 which conduct it by a small outlet to the channel on the other 
 side." 
 
 In the figure (Fig. 43) the left-hand filter is represented as un- 
 screwed and with two of the cloths raised for purposes of clean- 
 ing. The cloths are readily removed and replaced by a fresh set as 
 often as may be necessary — how often depends upon the amount 
 and the character of the impurity, other than the chemically 
 formed chalk, present in the water. At Liverpool and other 
 places where the waters are from deep wells in the chalk or red 
 sandstone, the filters run for 15 hours without changing the 
 cloths and the labor of one man is found sufiScient to cleanse 
 cloths, and filters and to attend to other details of the process in 
 softening 150,000 to 180,000 (imperial) gallons for the day's work. 
 In other places, where the water contains a larger proportion of 
 magnesia or surface impurities, the filters may not run for more 
 than 6 or 7 hours without cleansing. 
 
 From the application of Clark's process, in whatever form, 
 there results a large quantity of chalk or "whiting," more or less 
 pure according to the amount of impurity in the water softened. 
 If the water contains organic matters, a portion is precipitated 
 along with the chalk, together with an^ sediment which the 
 water may contain. In some localities there is a market for the 
 whiting which tends to offset the expenses of the process. A 
 portion might be burned into lime and used over again in soft- 
 ening a fresh portion of water, but being in a fine powder it could 
 not be burned in ordinary kilns. 
 
 Permanent hardness. — The permanent hardness is usually 
 caused by the presence of the sulphates (or other soluble salts) 
 of lime and magnesia, gypsum (sulphate of lime) being the most 
 common ; the action on soap is the same as that of the bicar- 
 bonates, which has been discussed under temporary hardness. 
 Water containing sulphate of lime may be softened by adding 
 carbonate of soda, and this is the method commonly employed 
 in the laundry. The chemistry of the process is this : when car- 
 
TREATMENT BY CHEMICAL PROCESSES. 187 
 
 bonate of soda in solution is mixed with sulphate of lime in solu- 
 tion, there are formed carbonate of lime (which settles out in the 
 sohd form) and sulphate of soda (which remains dissolved) ; a 
 similar action takes place with other soluble compounds of lime 
 and magnesia. The expense of this treatment makes it imprac- 
 ticable to soften, in this way, the entire water supply of a town, 
 a large portion of which is used for purposes where the hardness 
 of the water is a matter of indifference. We have seen (page 5) 
 that sulphate of lime becomes insoluble in water at high tem- 
 peratures and contributes to the formation of scale in steam- 
 boilers ; hence, for technical purposes, it is desirable to remove 
 the sulphate, and the process just indicated, or some other method, 
 maybe employed to advantage. 
 
 Treatment by other Chemical Processes, 
 
 On the large scale, attempts are seldom made to improve a 
 natural water except by processes which have already been de- 
 scribed ; there are, however, a number of substances which may 
 be used to advantage in purifying the small quantity necessary 
 for drinking in localities where no really drinkable water exists. 
 One of the longest used and best known substances is common 
 alum, which is often added to a turbid water. Where the water 
 contains carbonate of lime in solution, a chemical action takes 
 place between it and the alum, resulting in the formation of sul- 
 phate of lime, which remains dissolved, of carbonic acid, which es- 
 capes as gas, and of hydrate of alumina, which separates out as a 
 solid. As the hydrate of alumina forms and settles down, it entan- 
 gles and drags down with it the finely divided suspended matter 
 to which the turbidity is due : in fact, it enters into some sort of 
 chemical combination with some of the dissolved organic matter 
 which is thus removed. When the water does not contain 
 enough carbonate of lime or other substance capable of decom- 
 posing the alum, the addition of the alum may be followed 
 by the addition of a proper amount of carbonate of soda. Per- 
 chloride (or other soluble persalt) of iron acts very similarly to 
 alum. Instead of hydrate of alumina, it is the hydrate of iron 
 (ferric hydrate) which is formed, and this also, in settling, carries 
 with it some organic matter. It was at one time proposed to 
 treat the water of the Seine at Paris with alum, and the use of 
 
1 88 WATER SUPPLY. 
 
 perchloride of iron and carbonate of soda were talked of for ren- 
 dering potable the water of the Maias in Holland. 
 
 Another substance which has more recently come into notice 
 is permanganate of potash. The action is the same as that de- 
 scribed when discussing the use of this substance as a means of 
 determining analytically the amount of organic matter present 
 in a water. (See page 37.) In attempting to purify an impure 
 water by this means, the highly colored solution of the perman- 
 ganate must be added in quantity sufficient to impart a pink 
 color, which remains permanent for from five to ten minutes. 
 The permanganate, in destroying the organic matter, is itself 
 decomposed, and oxide (or hydrate) of manganese separates as a 
 finely divided solid. This may be removed by filtration, »r it 
 may simply be allowed to settle to the bottom of the tanks in 
 which the water is treated. Other permanganates may be used 
 as well as that of potash. For the treatment of the water which 
 the British army was likely to meet with upon the Gold Coast, 
 Professor Crookes,* in 1873, recommended a mixture of 
 
 I part of permanganate of lime, 
 10 parts of sulphate of alumina, 
 30 parts of fine clay. 
 
 He stated that this mixture, when added to London sewage in 
 the proportion of 20 to 10,000, afforded a very satisfactory puri- 
 fication. 
 
 The most simple manner of treating a water known or sus- 
 pected to be impure is to boil it, although it is by no means cer- 
 tain that immunity from harm is thus, in all cases, assured. 
 There is, however, evidence to show the value of the treatment ; 
 if, after the boiling, the water is iced, it becomes, of course, more 
 palatable. It is stated that the Chinese and Japanese drink no 
 water that has not been boiled ; and when we consider the un- 
 sanitary conditions which exist in those countries and the char- 
 acter of the water used, it seems as if boiling the water must 
 prevent ills that would otherwise befall the people. 
 
 In some instances lime has been added to water which is used 
 for domestic supply — the lime being added for purposes other 
 than the softening of a water containing the bicarbonates. Thus, 
 in Australia, at Sandhurst, Victoria, the impounded surface water 
 
 * Chemical News, xxviii (1873), p. 244. 
 
DISTILLATION. 
 
 189 
 
 which is used contains at times as much as 30 or 40 grains of 
 yellowish-brown clayey matter in the (imperial) gallon, i.e., say 
 about 50 parts in 100,000. Here it was found that sand filters 
 did not thoroughly intercept the clay in suspension, and that 
 the water, after filtration, still remained cloudy and opalescent. 
 Lime was added at the rate of 7 grains to the (imperial) gallon, 
 and after standing 10 hours the water became clear: five-sevenths 
 of the added lime went down with the precipitate, the other two- 
 sevenths remained in solution in the water, and of course gave 
 it a slight hardness. At several other works lime is used in the 
 ■same proportion, and this treatment is, in some cases, followed 
 by filtration, in others not. The capacity of one of the works 
 where this treatment is employed is as great as 1,000,000 imperial 
 gallons per day.* 
 
 Distillation. 
 
 The ordinary process of distillation is sufficiently familiar. 
 When water is boiled, the gaseous substances which it holds in 
 solution are expelled almost completely, either while the water 
 is heating up to the boiling point, or with the first portion of 
 the steam : the dissolved solids, on the other hand, remain be- 
 hind while the water evaporates. If the water be boiled in an 
 
 Fig. 46. 
 
 ordinary still — such, for instance, as is shown in Fig. 46 — and 
 the steam be subsequently cpndensed, the water which issues 
 from the condenser will be tolerably pure, especially if the first 
 
 * Brady : Proc. Inst. Civ. Eng. Gr. Br. Ivi, p. 134 and foil. 
 
ipO WATER SUPPLY. 
 
 portions be rejected and the evaporation be not carried too far. 
 Distillation is, of course, a somewhat expensive process, for be- 
 sides the cost of the necessary apparatus, each pound of water 
 evaporated requires the consumption of from one-twelfth to one- 
 seventh of a pound of coal, according to the quality of the coal 
 used and the efficiency of the boiler employed. On this account, 
 distilled water for drinking, cooking, and other domestic uses is 
 seldom prepared on any considerable scale except on shipboard. 
 Here, the steam is usually taken from one of the boilers used to 
 generate steam for the motive power of the ship, and the differ- 
 ent systems consist in differences in the condensers (aerators and 
 purifiers), although the term " distiller " is often applied to this 
 part of the apparatus. The condensers are of various forms — 
 the ordinary worm, the flat worm or zigzag, and the tube con- 
 denser, consisting of a cylinder with tubes inside running verti- 
 cally, the steam passing through the tubes and the water being 
 on the outside, or vice versa. 
 
 Ordinary distilled water has a fiat and nauseous taste, owing 
 partly to the fact that the dissolved gases, notably oxygen and 
 carbonic acid, have been expelled, and partly to the presence of 
 certain volatile organic compounds which have been formed 
 during the distillation. This is often remedied by allowing the 
 water to remain for from lo to 15 days in partly filled tanks 
 where it will Be exposed to the air and more or less agitated by 
 the motion of the ship. A number of devices have been con- 
 trived and patented which aim to accomplish this aeration and 
 oxidation at once, and thus to produce a distilled water fit for 
 immediate use. 
 
 In the United States navy, a condenser invented by Passed 
 Assistant Engineer G. W. Baird is largely used. The essential 
 feature of this system is the introduction of air into the apparatus 
 in such a way that it mixes with the steam, and the water which 
 forms is condensed in the presence of an abundance of oxygen, 
 and thus becomes fully aerated. It is further claimed that the 
 air so introduced oxidizes the organic matter carried forward by 
 the steam. This it no doubt does, if not at once, at least when 
 the water, thus aerated, is subsequently passed through a filter 
 of purified animal charcoal. Fig. 47 represents the condenser in 
 section. The steam enters at a and, on the principle of an 
 injector, draws air in through b ; the mixture of air and steam 
 
DISTILLATION. 
 
 191 
 
 enters the system of cooling pipes B, where the steam is con- 
 densed. The pipes are usually of cylindrical section and are 
 
 Fig. 47. — Baird's CoNDKNbtK. 
 
 made of tin or tinned copper ; they are coiled into helices, the 
 ends terminating in the common T-heads, C, C, at the upper and 
 lower ends respectively. The refrigerating water enters at d and 
 is discharged at e, and the condensed water flows out at f and 
 thence passes to the filter. In the U. S. navy the water is not 
 salified and is regarded as perfectly wholesome. It is stated * 
 that in the Russian navy there is added to each 1,000 liters of 
 distilled water a mixture consisting of 4.8 grams salt, 3.4 grams 
 
 * Fonssagrives : Hygiene et assainissement des villes ; Paris, 1874, p. 316. 
 
192 
 
 WATER SUPPLY. 
 
 sulphate of soda, 48 grams bicarbonate (sic) of lime, 14 grams 
 bicarbonate of soda, and 6 grams. carbonate of magnesia. • 
 
 The British navy uses Normandy's system, and essentially the 
 same apparatus is used in the German navy, as shown in Figs. 
 48 and 49.* The apparatus consists essentially of two cylinders, 
 the two sectional views of B being taken at right angles to each 
 other. Steam is generated in a separate boiler and enters the 
 cylinder A through the pipe d, and passes into the sheaf of tubes 
 
 b which are surrounded 
 
 nby water which is to be 
 distilled. The water 
 
 Fig. 48. Fig. 49. 
 
 formed by condensation collects in the reservoir e at the base 
 of the sheaf of pipes, and from here it flows into the vessel g, 
 shown only in Fig. 48 ; if the water is required warm, it may bt 
 
 * Fischer : Chemische Technologie des Wassers p. 205 and foil. 
 
DISTILLATION. Iq, 
 
 drawn from g directly, otherwise it flows through a connecting 
 pipe into the lower sheaf of tubes in the cylinder B. The level 
 of the water in A— which flows in from B— is so regulated that 
 the steam which is formed may free itself from any salt water 
 which it carries mechanically, by passing first through the per- 
 forated copper plate a, and then by striking against c, before it 
 passes through the pipe m into the sheaf of tubes n in the upper 
 part of the cylinder B. The water which is here condensed flows, 
 for further cooling, into the lower sheaf of tubes in B, into which 
 the water condensed in b also flows, and which is the set of tubes 
 v/ith which the cooling water entering at i comes in contact. 
 The water, thus fully cooled, is either discharged directly through 
 the tube r, or is conducted through .r to a simple filter filled with 
 animal charcoal. The water used in cooling enters through the 
 tube i and flows off through k, except so much as is necessary to 
 keep the level of the water in A at the proper height. The air 
 which escapes from the water as it becomes warm is conducted 
 through the tube t into the steam space of the distilling ap- 
 paratus, in order that the water, as it condenses, may dissolve it 
 again and .thus become more palatable. By means of the con- 
 nection jf the return steam from the steam pump may be allowed 
 to enter ^, to be utiHzed in the production of distilled water. 
 Finally, when the water in the distilling apparatus. A, becomes 
 too concentrated, it may be withdrawn by means of a cock not 
 shown in the figures. 
 
 While distilled water is seldom prepared on a large scale ex- 
 cept on shipboard, there are some localities where drinking water 
 cannot otherwise be procured, and distillation must be resorted 
 to. For example, the island of Walcheren, in Holland, is de- 
 pendent for its supply of fresh water for drinking on the rainfall, 
 all other water being brackish. For the supply of ships leaving 
 the harbor of Flushing (Vlissingen), it has been necessary to have 
 recourse to the condensation of steam, as being the only avail- 
 able source of fresh water independent of rain; Normandy's ap- 
 paratus is employed. It is stated * that the plant cost 20,000 
 Dutch florins (about $12,000), and that 18 kilograms of distilled 
 water are produced for each kilogram of coal burned, the water 
 being distilled at the rate of one cubic meter per hour and being 
 of satisfactory quality. 
 
 * Prcc. Inst. Civ. Eng. Gr. Br. , Ixii, p. 408. 
 13 
 
CHAPTER X. 
 
 SOME GENERAL CONSIDERATIONS. 
 
 Quantity and Waste. 
 
 Whatever source may be chosen for the supply of a city or 
 town, it is essential that the quantity of water should be suffi- 
 cient for the needs of the population for a number of years. 
 The experience of many places has been similar to that of New 
 York City. Within eight years after the completion of the Cro- 
 ton aqueduct, the New York Water Department wrote : " This 
 Board warns the City Council, and through it every citizen, that 
 every drop of water which the works in their present state can 
 supply is now being delivered in the city." What is, however, a 
 sufficient quantity ? The following table gives the amount of 
 water per head of population consumed in certain European 
 cities: 
 
 table XXVI. — Consumption of Water. in European Cities. 
 
 Cities (iB3o). ■* 
 
 Glasgow. . .. 
 
 Paris 
 
 ■Edinburgh. . 
 
 Dublin 
 
 Hull 
 
 Birmingham 
 Blackburn . . 
 Leeds. . . .'.. 
 Liverpool . . 
 Manchester . 
 Sheffield . . .' 
 
 Daily Average Supply 
 
 PER HEAD 
 
 U. S. Gallons. Litres. 
 
 60 
 
 50.2 
 
 48 
 
 45-6 
 
 39-2 
 
 30 
 
 30 
 
 27.9 
 
 27 
 
 24 
 
 21.6 
 
 227 
 iqo 
 181 
 172 
 148 
 
 "3 
 113 
 107 
 104 
 
 91 
 
 82 
 
 Cities, t 
 
 Karlsruhe 
 
 Bonn 
 
 Hamburg 
 
 Dresden 
 
 Frankfurt a. M 
 
 Coin 
 
 Altenburg .... 
 Braunschweig . 
 
 Bamberg 
 
 Kassel 
 
 Hannover 
 
 Altona 
 
 Leipzig 
 
 Daily Average Supply 
 
 per head, in 
 U. S. Gallons. Litres. 
 
 154 
 76 
 03 
 60 
 59 
 53 
 43 
 41 
 39 
 33 
 31 
 31 
 23 
 
 581 
 289 
 237 
 228 
 223 
 200 
 163 
 154 
 148 
 124 
 116 
 
 115 
 86 
 
 The estimates of European experts as to the amount of 
 water necessary for an adequate supply must be received with 
 
 "" Brackett : Journ. Assoc. Eng. Soc, I (1882), p. 261. 
 
 f Grahn : Journ . f. Gasbeleuchtung u . Wasserversorgung, x.x (1877). 
 
QUANTITY AND WASTE. 
 
 19s 
 
 some caution as applied to American circumstances, owing to 
 difference in a variety of conditions. The following table shows 
 the actual consumption in a number of American cities : •'■ 
 
 TABLE XXVI. — Consumption of Water in American Cities. 
 
 American Cities. 
 
 Fall River.. 
 Providence . 
 
 Lowell 
 
 Cambridge. . 
 Brooklyn . . . 
 Philadelphia 
 St. LoiiiB . . . 
 Cincinnati . . 
 New York . . 
 
 Boston 
 
 Chicago .... 
 Detroit 
 
 Year. 
 
 1880 
 l88l 
 1880 
 
 i87g 
 i8^o 
 
 iSSi 
 1880 
 
 iSSi 
 
 Population. 
 
 49,430 
 102,500 
 
 59,485 
 52,B8o 
 566,689 
 847,542 
 346,000 
 256,708 
 1,206,590 
 416,000 
 503.304 
 iiS,ooo 
 
 Average Daily 
 
 Gallons 
 
 Consumption. 
 
 per head 
 
 Gallons. 
 
 per day. 
 
 1.448,247 
 
 30.1 
 
 3.716,937 
 
 36.3 
 
 2,252,197 
 
 37.9 
 
 2,472,108 
 
 46.7 
 
 30,744,590 
 
 54.2 
 
 57,707,082 
 
 68.1 
 
 24,958,000 
 
 72.1 
 
 19.476,739 
 
 75-9 
 
 95,000,000 
 
 78.7 
 
 38,214,900 
 
 92. 
 
 57.384,376 
 
 114. 
 
 17,926,377 
 
 151.9 
 
 Litres 
 per head 
 per day. 
 
 114 
 137 
 143 
 177 
 205 
 
 257 
 273 
 287 
 297 
 34S 
 431 
 574 
 
 From this table it is evident that there is a great difference 
 in the amount of water consumed in different places, and if from 
 30 to 50 gallons, suffice in certain cities, the use of 90 or 100 
 gallons in others presupposes a considerable waste : in fact, it is 
 generally agreed by those in charge of water supplies that from 
 a quarter to one-half of the water furnished is actually wasted. 
 Mr. Thos. J. Whitman, of the St. Louis water works, said, a few 
 years ago, that it cost the city fully $300,000 annually in fuel 
 alone to simply supply the waste, and similar statements come 
 from all large cities. For domestic and household uses 20 gal- 
 lons per person per day is a sufificient allowance : taking into 
 account the water used for manufacturing and mechanical pur- 
 poses, that necessary for street sprinkling, extinguishing fires, 
 for use in stables, etc., 60 gallons per day for each inhabitant is 
 a liberal quantity in the case of large cities and manufacturing 
 towns. In the case of the smaller, non-manufacturing towns 35 
 or 40 gallons should suffice. Mr. Dexter Brackett, in a valuable 
 paper on the waste of water, from which Table XXVI and a 
 part of Table XXV have been taken, considers 50 gallons per 
 head as sufificient to provide for all 'the demands of the largest 
 cities of the country. 
 
 The great waste which takes place being acknowledged, the 
 
 * Brackett : Journ. Assoc. Eng. Soc. , I, p. 261. 
 
196 WATER SUPPLY. 
 
 question arises how to prevent, or at least diminish it. There 
 are two general methods which suggest themselves. The first is 
 a rigid system of inspection in order to detect all leakage from 
 the pipes and from imperfect fixtures, as well as all unlawful use 
 of the water ; the second is the general and compulsory intro- 
 duction of meters. With reference to the first method, although 
 the authorities in charge of the distribution of water reserve the 
 right to inspect the fixtures at any time, yet such inspection is 
 annoying and repugnant to the average householder, and the 
 system admits of abuse. A modification of this system, which 
 obviates the necessity of entering the. building except in cases 
 where abnormal use of water is already known to exist, is that 
 devised by Mr. G. F. Deacon, Borough Engineer, Liverpool, 
 England."" The following description of the system is taken 
 from a report on waste of water made to the City Council of Bos- 
 ton, Mass.f 
 
 " In the Deacon system the waste-water meter is used to locate 
 sources of waste. This meter does not, like the ordinary meter, 
 record the number of gallons consumed, but it indicates the rate 
 of flow at any given time, and whether the discharge is due to 
 steadily flowing waste, or to intermittent and ordinary use. It 
 therefore enables the observer to determine, by observations 
 taken at those hours when no water, or a very small quairtity, is 
 used for legitimate purposes, whether waste is going on. 
 
 " The meter (Fig. 50) 'j: consists of a hollow cone, having its 
 small end upwards, and containing a composition disk, of the same 
 diameter as the small end of the cone. A vertical spindle, attached 
 to the upper surface of this disk, is suspended by a fine German- 
 silver wire, which passes, practically water-tight, through a small 
 hole in the top of the chamber, over a pulley, and supports a 
 weight. This weight is so adjusted as to retain the disk at the 
 top of the cone when the water is at rest. When any water is 
 drawn through the meter, the disk is pressed downward towards 
 the bottom of the cone, its position depending upon the amount 
 
 * A valuable report of Mr. Deacon is reprinted in the Report of the Cochituate 
 Water Board of the City of Boston, for the year ending April 30, 1874. City Docu- 
 ment No. 55, pages 84-112. 4 
 
 t Report on Waste of Water (May 2^, 1882). Boston City Document No. 78,1882. 
 
 it Figure 50 is reduced from a plate in a paper on The Constant Supply and. 
 Waste of Water, by Geo. F. Deacon, in the Journal of the Society of Arts, May, 1882. 
 
FOUR INCH DEACON WASTE WATER METER. 
 
 ^v,.^^k.k^v^^k^vvv^'\v^<<<^\ttt\^^^^^^ 
 
 A Inlet. 
 
 B Outlet. 
 
 C Gauge Cone. 
 
 E Stem of Disk. 
 F Guide for Stem. 
 
 G Wire connecting 
 i Disk with Pen- 
 
 cil Carriage. 
 H Gland with 
 
 Fig. 50. 
 
198 
 
 WATER SUPPLY. 
 
 of water passing through the meter. 
 
 By means of a pen- 
 2 cil attached to the 
 wire the motions of 
 the disk are record- 
 ed on a drum, which 
 revolves by clock- 
 work, once in 24 
 hours. 
 
 " A fac-simile, 
 about one-fifth 
 full size, of a dia- 
 gram drawn auto- 
 matically by a 
 waste-water meter, 
 is shown in Fig. 51. 
 It is obvious that 
 when water is being 
 drawn off for use 
 the rate of flow 
 from minute to 
 minute must be va- 
 riable ; and this is 
 accordingly shown 
 by the irregular 
 vertical lines from 
 noon to midnight, 
 and from 4 A. M. to 
 noon. When con- 
 t i n u o u s — that is, 
 preventable, waste 
 alone is takingplace 
 — the flow must evi- 
 dently be uniform ; 
 and this condition 
 is indicated by the 
 comparatively uni- 
 form and horizontal 
 line from i to 4 
 
 A.M., only occasionally broken by vertical lines, caused by persons 
 
 drawing water during the night. 
 
QUANTITY AND WASTE. igg 
 
 "The meter is placed in a box under the sidewalk or roadway, 
 and so located as to control the flow of water supplied to a cer- 
 tain district, the limits of which have been previously determined. 
 All the water used in this district is drawn through the meter, 
 and the quantity and rate recorded. After a few diagrams have 
 been taken, to show the ordinary rate of consumption, inspection 
 is commenced. Every service pipe is provided with a stopcock, 
 which is accessible from the sidewalk by means of an iron wrench 
 about seven feet long. When this wrench is applied to the stop- 
 cock the sound caused by water passing through the service pipe 
 can be easily distinguished. When no noise is heard, with the 
 stopcock fully open, it is partly closed, and the increased velocity 
 always causes a distinct sound, although the quantity of water 
 passing the stopcock may be very small. A night inspector be. 
 gins his work about midnight, and tests, by means of his shutting- 
 off wrench, each service pipe. If he discovers any flow through 
 the service pipe, the stopcock is closed, and a note made of the 
 time and the number of the house. He continues this operation 
 through the district until about 4 A.M., when he retraces his 
 steps, and opens all the stopcocks he had found wasting. Dur- 
 ing this time the meter is recording the consumption, and the 
 diagrams show the amount of water wasted by each of the ser- 
 vice pipes that were closed, the time the inspector began and 
 finished his work, and the time each stopcock was closed. The 
 day inspector receives the night inspector's report, visits the 
 premises where waste was noted, and ascertains the cause. In 
 cases of waste from defective fixtures the owners are notified to 
 repair the same, and' the visits are continued until the notices 
 have been complied with. 
 
 " The economy of this system, as compared with house-to- 
 house inspection, is apparent. The attention of the inspector is 
 at once directed to the place where the waste is going on, and 
 the time lost in indiscriminate inspection is saved." 
 
 The Deacon system has given very satisfactory results where- 
 ever tried. The chairman of the Liverpool Water Committee, 
 in his, address in 1879, ^^^^ ' 
 
 "We have given the city a constant service* of water, with a 
 
 * Before the introduction of the Deacon system of inspection, the supply of Liv- 
 erpool was an intermittent one, the water being on only gj hours out of the 24. 
 
200 
 
 WATER SUPPLY. 
 
 decline in the death-rate, and it now remains for me to show 
 what other effects have arisen from the change of system. In 
 the year 1871 we delivered an average of 122,000,000 gallons 
 weekly ; in 1880 our deliveries will be about 11 5, 000,000- gallons; 
 and this, notwithstanding an increased sale for trade purposes 
 of 12,250,000 gallons weekly, and an increased population of 
 104,000. The saving has been so great as to meet the increasing 
 demands of the city and district for eleven years. 
 
 "The change to constant service has already yielded nearly a 
 quarter of a million of money (;^250,ooo), and the ultimate saving 
 to the rate-payers, when the 7,000,000 gallons per week yet un- 
 sold are absorbed, will be £e,o,ooo per annum." 
 
 The conditions under which this system of waste detection 
 has been tried in Glasgow correspond more nearly with those 
 existing in Boston and other American cities, than do those of 
 other European cities where the system has been used. The 
 supply furnished is constant and ample, and the proportion of 
 water fittings to the population is larger than is common in most 
 European cities. 
 
 The following table (Brackett) shows the results obtained in 
 Glasgow from an inspection similar to the one made in Boston : 
 
 TABLE XXVII. — Waste Water Inspection in Glasgow. 
 
 J) 
 
 i 
 
 
 ii 
 
 Gallons per Head per Day. 
 
 
 s 
 
 ag 
 
 p. H 
 
 
 
 
 
 
 ' SB 
 
 '4 
 
 25 
 
 Sg 
 
 At starting of meters. 
 
 After first three inspec- 
 tions. 
 
 n 
 
 li 
 
 W „ M 
 
 
 
 
 Nig^ht rate 
 
 
 Night rate 
 
 s 
 
 1 
 
 D Pt 
 
 gss 
 
 Total. 
 
 per 
 
 Total. 
 
 per 
 
 a 
 
 ^; 
 
 Z 
 
 S 
 
 
 24 hours. 
 
 
 24 hours. 
 
 I 
 
 9 
 
 14,972 
 
 25.8 
 
 71.0 
 
 54.0 
 
 40.9 
 
 21. 1 
 
 11 
 
 6 
 
 10,002 
 
 22.4 
 
 79.0 
 
 72.2 
 
 50.4 
 
 33-0 
 
 III 
 
 3 
 
 4,g86 
 
 35-4 
 
 73-7 
 
 62.4 
 
 44.2 
 
 21.7 
 
 IV 
 
 6 
 
 7,629 
 
 30.6 
 
 79.0 
 
 57-0 
 
 50.5 ' 
 
 24.8 
 
 V 
 
 7 
 
 9.815 
 
 37-8 
 
 55-1 
 
 36.8 
 
 37-7 
 
 17-3 
 
 VI 
 
 8 
 
 12,614 
 
 39.7 
 
 37-2 
 
 29.5 
 
 27.1 
 
 17. ,8 
 
 VII 
 
 Z 
 
 4,132 
 
 25-5 
 
 4-5-5 
 
 30.6 
 
 41.9 
 
 25.8 
 
 VIII 
 
 3 
 
 6,306 
 
 37-1 
 
 44.9 
 
 27.5 
 
 33-6 
 
 15-1 
 
 IX 
 
 4 
 
 7,821 
 
 32. i 
 
 44.9 
 
 31-7 
 
 30.8 
 
 15-2 
 
 X 
 
 2 
 
 3,012 
 
 34.6 
 
 44.2 
 
 33-4 
 
 25.0 
 
 12.8 . 
 
 Totals and 
 averages. 
 
 
 
 
 
 
 
 
 50 
 
 81,289 
 
 30.6 
 
 55.8 
 
 45.2 
 
 38.4 
 
 21.0 
 
 The results of a trial of the Deacon system on a portion of 
 
QUANTITY AND WASTE. 
 
 20 1 
 
 the water service of the city of Boston, Mass., are shown in the 
 following table (Brackett). 
 
 ■ TABLE XXVIII.- 
 
 -Waste Water Inspection in Boston, Mass. 
 
 
 !; * - 
 
 
 ii 
 
 Gallons per Head per Day. 
 
 
 
 ■) 
 
 
 H "1 K 
 
 :^ S u 
 
 
 
 
 Percentage 07 
 Reduction. 
 
 ';'■ Number 
 
 ; I OF 
 
 § 
 
 IS 
 
 Before inspection. 
 
 After two or three 
 inspections. 
 
 Section. 
 
 Total. 
 
 Night 
 rate per 24 
 
 Total. 
 
 Night 
 rate per 24 
 
 On total. 
 
 On night 
 
 ''• J., 
 
 u 
 
 ^ 
 
 
 hours. 
 
 26.4 
 
 hours. 
 10.6 
 
 
 
 \ >'■ I 
 
 2,810 
 
 9.2 
 
 53-5 
 
 39-1 
 
 50-7 
 
 72.9 
 
 I and I A 
 
 3,675 
 
 91 
 
 52- 
 
 39.0 
 
 34-1 
 
 13-7 
 
 34-4 
 
 64 
 
 Q 
 
 2 
 
 2,170 
 
 8.x 
 
 49-9 
 
 33 I 
 
 36.7 
 
 13.2 
 
 26.5 
 
 60 
 
 6 
 
 3 
 
 2,030 
 
 6.2 
 
 71.8 
 
 43.2 
 
 45-' 
 
 20.2 
 
 37.2 
 
 53 
 
 2 
 
 4 
 
 i,8So 
 
 5-9 
 
 68. 4 
 
 42.2 
 
 47 8 
 
 22.3 
 
 30.1 
 
 47 
 
 2 
 
 5 
 
 1,790 
 
 5-9 
 
 72.7 
 
 53-3 
 
 47.8 
 
 17.8 
 
 34-3 
 
 66 
 
 6 
 
 6 
 
 1,875 
 
 7-2 
 
 60. 
 
 44.6 
 
 35-3 
 
 15-1 
 
 41.2 
 
 66 
 
 I 
 
 7 ■ 
 
 2,540 
 
 6.8 
 
 55-2 
 
 31-9 
 
 39-6 
 
 19.2 
 
 28.3 
 
 39 
 
 8 
 
 8 
 
 2,400 
 
 6.6 
 
 55- 
 
 40. S 
 
 37-9 
 
 18.5 
 
 3I-I 
 
 54 
 
 7 
 
 1 '' -' 
 1 \' 11 
 
 f . 12 
 
 2,150 
 
 5-5 
 
 62.9 
 
 40.1 
 
 36.2 
 
 13-7 
 
 42.4 
 
 65 
 
 8 
 
 1,790 
 
 5-6 
 
 52-3 
 
 28.1 
 
 46. 1 
 
 18.7 
 
 II. 9 
 
 33 
 
 4 
 
 2,800 
 
 6.5 
 
 43-7 
 
 17-5 
 
 25-7 
 
 9-5 
 
 41.2 
 
 45 
 
 7 
 
 2,300 
 
 7.6 
 6.86 
 
 80.4 
 
 55-2 
 
 31.2 
 
 
 
 37-7 
 
 12. 5 
 15-8 
 
 6l.2 
 
 77 
 
 4 
 
 '< Averages 
 
 
 53. 5 
 
 37-5 
 
 35-6 
 
 57 
 
 9 
 
 From the above it appears that on the whole district covered 
 by the inspection, containing a population of 21,760 persons, the 
 average daily consumption was reduced from 58.5 to 37.7 gallons, 
 a saving of 35.6 percent, or 20.8 gallons for each person supplied, 
 while the night rate was reduced from 37.5 to 15.8 gallons per 
 head per day, a saving of 58 per cent. The Boston experiments 
 being continued for a short time only, seem to have cost more 
 than the value of the water saved ; this is not the case where the 
 system forms a part of the regular operation of the water works. 
 
 The Cincinnati Board of Public Works use a device invented 
 by Mr. Thomas J. Bell, the Assistant Superintendent of the 
 Water Works, which takes advantage of a fact long known, 
 namely, that it is possible to detect a leak by taking advantage 
 of the conduction of the sound caused by that leak through the 
 metal pipes by some metallic connection, to the ear. Mr. Bell's 
 device consists of a diaphragm inclosed in a hollov/ piece of 
 wood, shaped like a telephone, and about the same size. A piece 
 of iron extends through the middle of the neck of the " detect- 
 or," one end projected and threaded, and the other communicat- 
 ing with the diaphragm. A threaded hole is bored in the top 
 
202 
 
 WATER SUPPLY. 
 
 of the key used to turn water on and off at the cocks on the edge 
 of the pavements, and, when it is desired to make a test, the de- 
 tector is screwed into the top of the key and the ear appHed to 
 the detectcr. The least leakage or the smallest stream running 
 from a hydrant can be distinctly heard. Inspections are made at 
 night. 
 
 A waste-water indicator, invented by Mr. Benj. S. Church, 
 resident engineer of the Croton aqueduct, N. Y., is thus de- 
 scribed in The Sanitary Engineer, to the publisher of which we 
 
 are indebted for the use 
 of the cut (Fig. 52). 
 
 " The device consists 
 of a pressure gauge, with 
 arrangements for attach- 
 ing it to the pipe through 
 which water is suspected 
 of leaking. Its most ex- 
 tensive application is de- 
 signed to be to service 
 pipes from street mains 
 to houses. For this pur- 
 pose, a special stop-cock 
 is placed on the service 
 pipe under the sidewalk 
 instead of the ordinary 
 kind, and from it a pipe 
 or hollow stem, instead of 
 a solid rod, runs up to the 
 stop-cock box. The shank 
 of the key, also, is hollow, 
 and has a pressure gauge 
 attached at the top. By 
 means of a coupling, 
 turned by a wheel at its top (shown just below the handles of 
 the key on either side of the gauge), an air tight connection is 
 made between the stem and the key. When the cock is closed 
 (by turning the cross-arms, and with them the gauge, which is 
 fastened to them), a small ' port ' in the stopcock plug establishes 
 a connection between the street main and the gauge ; the air of 
 the stem is compressed, and the hand on the gauge indicates the 
 
 Fig. 52. — Church's Detector. 
 
QUANTITY AND WASTE. 
 
 203 
 
 hydrostatic pressure on the main. By turning the plug to another 
 position, a second ' port ' is opened, allowing the water to flow 
 through, if there be any escape from the pipes in the house. If 
 that be the -case, the gauge will indicate a diminished pressure, 
 depending on the amount of the discharge. The plug can also 
 be turned to a third position, leaving it open to the house and 
 closed against the street. By this means the approximate height 
 of the leak (if such is found) is ascertained. The exact position 
 of the ports is indicated on a scale with vernier, which slides up 
 and down the coupling, to be clamped at the level of the side- 
 walk for convenience in making reference marks thereon. The 
 gauge is said to indicate a flow of five gallons an hour, and ' re- 
 veal the least leakage even to the size of a pin-hole.' The ap- 
 paratus may also be applied to street mains to detect leaks in 
 them." * 
 
 The second general method for checking waste is the universal 
 introduction of meters. There can be no question of the justice 
 and propriety of measuring the water used for manufacturing 
 purposes or in hotels and other large establishments. It is, how- 
 ever, objected that the adoption of meters for private dwellings 
 will cause an injurious economy in the use of water among the 
 very class of the population where it is important that water 
 should be used freely. This objection is obviated to a great 
 extent, at least, in some places where meters are employed, by 
 fixing a minimum tax — to be paid by all water-takers — which 
 shall cover a certain quantity, based on a reasonable estimate of 
 domestic needs. Water used in excess of this quantity is paid 
 for by measurement, and special arrangements may be made for 
 tenement houses and the dwellings of the very poor. It is 
 further said that the cheaper meters are not very reliable, that 
 it is often possible to pass a considerable amount of water with- 
 out its being registered, provided the water be allowed to flow 
 slowly, and that the meters are continually needing repairs and 
 giving rise to dissatisfaction and complaint on the part of the 
 water-takers. 
 
 Meters are in very general use in Providence and Pawtucket, 
 R. I., and in Fall River, Mass., having been introduced with the 
 water supply: Providence, with 9,780 services has 4,816 meters, 
 and consumes 36.3 gallons per head of population ; in Fall River 
 
 * The Sanitary Engineer, Januaiy i8, 1883. 
 
204 WATER SUPPLY. 
 
 60 per cent, of the services are metered, and the consumption is 
 30.1 gallons per head. The introduction of meters is, of course, 
 much more easy with new works than with works already estab- 
 lished. No statistics have come under the author's observation 
 with reference to the matter of repairs in those American cities 
 where the use of meters is anything like general. The following 
 statistics are compiled from German sources : 
 
 In Magdeburg it was decided in 1879 t° introduce meters 
 universally, and the water rate was fixed at 33 marks (say $8.00) 
 per house for any amount up to 300 cubic meters per annum 
 (about 215 U.S. gallons per day). -Above this amount the 
 charge was made at the same rate, i. e., 11 pfennige per cubic 
 meter. The introduction of meters was decided upon almost as 
 a necessity, but, according to subsequent official reports, has 
 been in every way successful. At the close of the year 1880 
 there were 2,792 meters in use. During the year 168 meters 
 had required repairs, and yo had been tested at the request of 
 consumers. The meters are all tested before introduction, and a 
 , record kept for future comparison. In Berlin meters are in uni- 
 versal use. Eighty-two per cent of the water supplied is 
 measured and paid for ; the remainder includes what is used for 
 flushing the pipes, for extinguishing fires, for sprinkling streets, 
 etc., and also loss by leakage. Of the 15,853 meters in use in 
 1880-81, 2,093, or 13.2 per cent, were removed for more or less 
 serious defects, the number registering incorrectly or not at all 
 being 1,580, or ten per cent of the whole number in use. In 
 Breslau the number of meters in use in 1880-81 was 5,141 ; of 
 these, during the year, 1,085 were tested by request of the con- 
 sumer or at the instance of the inspector. Of the 1,085 tested, 
 740 were found to need repairs of some sort, 406 {i. e., 7.9 per 
 cent) registered incorrectly or not at all. It should be said that 
 the error in registering is generally to the benefit of the con- 
 sumer. 
 
 Of the methods indicated to check waste, local considerations 
 must determine which shall be adopted. If a city " has at its 
 disposal 150 million gallons daily for a population which does 
 not consume 30 miUions," as is stated to be now the case in Balti- 
 more, Md., there is certainly no occasion for introducing meters 
 at all ; but in places where the available water is limited in 
 quantity, or where, without economy, the existing supply is 
 
QUANTITY AND WASTE. • 205 
 
 likely to prove insufficient in the immediate future, or in places 
 where the rates are of necessity high, the introduction of meters 
 would be advisable. The waste in Northern cities during the 
 winter is enormous, as it is very common to leave the faucets 
 open during the night, in order to prevent freezing. This cold 
 weather waste can never be completely stopped until property 
 owners are obliged to arrange their plumbing so that the water 
 can be completely drawn from the pipes when liable to freeze. 
 By the use of meters it can be largely reduced, as the waste 
 would be reduced to the minimum amount required to prevent 
 the pipes from freezing, and it would become a question to the 
 water-taker whether it was economy to waste water or remodel 
 his fixtures. 
 
 Next in importance to the special form of waste just alluded 
 to, is that due to defective or improper fixtures. Such sources 
 of waste may be readily detected by systematic inspection, 
 and should be controlled by municipal regulation, as is indeed 
 done in some cities. " Providence, New York, Brooklyn and 
 other American cities license their plumbers, and to a certain 
 extent inspect the fixtures used ; but in English cities the ordi- 
 nances and regulations are much more rigid than those in this 
 country. Liverpool, Manchester, Glasgow and other English 
 cities test and stamp all of the water fittings used. In Glasgow, 
 during the year 1877, when this plan was first adopted, of 4,369 
 fittings examined, 14.6 per cent were rejected, while in 1880, of 
 27,517 examined, but 3.92 per cent were rejected. In the latter 
 city certain varieties of fittings are proscribed. When any of 
 these are found wasting water twice during three months, they 
 are removed, and their use is not allowed at all in new prem- 
 ises. 
 
 " All cisterns are provided with overflow pipes, which are 
 brought outside the building or made to discharge inside where 
 they can be seen. The service pipes, except in special cases, are 
 required to be of lead, aind their weight is prescribed. No pipe 
 or fitting can be covered until inspected, to see that it conforms 
 to the regulations. 
 
 " Water-closets and urinals are not allowed to be supplied 
 direct from the service pipe, but must be supplied from cisterns, 
 so constructed, that in water-closets not more than two gallons 
 can be used at a single flush, and in urinals not more than li 
 
206 WATER SUPPLY. 
 
 gallons, and so that they cannot be made to flow continuously 
 either by intention or neglect. 
 
 " The adoption of the above or similar regulations in Ameri- 
 can cities, while not in the least curtailing the legitimate use of 
 water, would be the means of preventing a very large proportion 
 of the present enormous amount of willful and useless waste." '•'' 
 
 In view of the difficulty of supplying large cities with water 
 against which no complaints can arise, the question is often 
 raised whether it is . not advisable, in some cases, to introduce a 
 double supply. As the author has remarked elsewhere : f "It is 
 true that for many purposes, as for extinguishing fires and for 
 sprinkling streets, a water would answer which would not be 
 suitable for drinking, and such a supply might in many cases be 
 easily procured, while to procure an abundance of water well 
 suited for drinking would involve a large outlay. To the double 
 system there is no (sanitary) objection, if the poprer water can 
 be drawn only from street hydrants, which are under municipal 
 control ; but it is not practicable to supply two sorts of water to 
 private dwellings, with any security that the distinction between 
 them- will be regarded ; no domestic, and indeed no average in- 
 habitant, will fail to use for all purposes that water which is most 
 handily obtained, unless, indeed, it be actually repulsive to the 
 taste." It should be said, moreover, that the introduction of a 
 second (inferior) water, where works already exist, would often 
 prove nearly or quite as expensive as the extension of the exist- 
 ing works and the increase of the supply of water fit to drink. 
 
 Conduits and Distribution Pipes. 
 
 Where the source of supply is at a considerable distance, the 
 water is usually carried by gravity in brick or other masonry con- 
 duits to a storage or distributing reservoir. Open canals are dis- 
 advantageous, especially on account of the liability of pollution, 
 but also as giving opportunity for considerable changes of tem- 
 perature and for vegetable growths. The water in passing 
 through a long conduit has some action on the mortar or cenient, 
 and may become slightly harder ; generally, however, the volume 
 of water is so great that there is very little perceptible effect. 
 Except in this respect, the water undergoes almost no change. 
 
 * Brackett, loc. Hi. f Buck's Hygiene, vol. i. p. 215. 
 
CONDUITS AND DISTRIBUTION PIPES. 207 
 
 Even the change of temperature in a properly covered conduit 
 with constant flow is small, Thus, Kerner found, in the hottest 
 days of the summer of 1875, that the water of the Frankfort 
 supply increased in temperature from 9° C. to only 10° C, in 
 flowing from the source to the main reservoir, a distance of' 82 
 kilometers. In passing through the city mains, a further distance 
 of six kilometers, the temperature increased 2^.75 C* (Compare 
 page 93.) Where the source of supply is a river or pond, a con- 
 siderable growth of the fresh-water sponge is often found on the 
 walls of the conduit for some distance. For this and other 
 reasons, such conduits should be subjected to periodical cleansing, 
 if inspection shows it to be necessary. 
 
 The distances of the sources of supply of various cities is as 
 follows : 
 
 Altenburg (springs) 15.5 miles. Munich (springs) 24.9 miles. 
 
 Danzig (springs) 12.4 " Paris (River Dhuis) 814 " 
 
 Franlifovt u. M. (springs). .. 52. " Paris (River Vanne) 107.2 " 
 
 Glasgow (kite) 36. " Vienna (springs) 60.3 " 
 
 Gotha (springs) 20. 5 " 
 
 Boston (Lake Cochituate). . . 16. " New Yorlc (Croton River) 40.6 " 
 
 From the distributing reservoir, or directly from the source of 
 supply, the water passes into the main distribution pipes, which 
 are usually of cast iron. Although there are some waters which 
 experience has shown to have almost no action on cast iron, with 
 most waters the pipes soon begin to rust. The rust often begins 
 at numerous isolated points, or nuclei, forming "tubercles," 
 which increase in size, become merged together and finally — aided 
 by the collection of sediment from the water — nearly choke the 
 pipe. The presence of iron in the water, either in solution or in 
 suspension, can hardly be regarded as deleterious to health, but 
 the water is sometimes rendered unfit for washing and cooking. 
 The presence and growth of the deposit of iron-rust, has, how- 
 ever, a very serious effect on the flow of the water, and it be- 
 :omes necessary to remove the deposit by scraping the inside 
 3f the pipes. Several special tools have been devised for this 
 Purpose. The following tables (Tables XXIX and XXX) will 
 ihow the effect of the accumulation of rust in diminishing the 
 
 * WolffhUgel : Wasserversorgung, p. 227. 
 
2o8 
 
 WATER SUPPLY. 
 
 capacity of the pipe and the flow of water : they are the results 
 of observations made in Aberdeen, Scotland.'" 
 
 TABLE XXIX. — Diminution of Available Capacity of Corroded Pipes. 
 
 
 
 
 Internal 
 Diameter. 
 
 Amount of 
 
 Capacity of 
 
 
 
 
 
 Rust per line- 
 
 CLEAN Pipe per 
 
 Percentage of 
 
 No. 
 
 Ace of Pipe. 
 
 Character. 
 
 ar Yard. 
 
 linear Yard. 
 
 Space occupied 
 
 
 
 
 
 
 
 DY Rust. 
 
 
 Years. 
 
 
 Inches. 
 
 Cu. Inches. 
 
 Cu. Inches. 
 
 
 I 
 
 20 
 
 Uncoated. 
 
 3 
 
 63.84 
 
 254.44 
 
 25.0 
 
 2 
 
 29 
 
 (C 
 
 3 
 
 86.94 ■ 
 
 254-44 
 
 34-1 
 
 3 
 
 38 
 
 
 3 
 
 110.44 
 
 254-44 
 
 43.4 
 
 4 
 
 29 
 
 " 
 
 4 
 
 182.37 
 
 452.37 
 
 40-3 
 
 5 
 
 22 
 
 " 
 
 4 
 
 244-37 
 
 452.37 
 
 54.0 
 
 6 
 
 14 
 
 " 
 
 5 
 
 180. 
 
 706.86 
 
 25-4 
 
 7 
 
 15 
 
 Coated. 
 
 7 
 
 igo. 
 
 1,385.42 
 
 13-7 ■ 
 
 8 
 
 15 
 
 '* 
 
 10 
 
 240. 
 
 2,827.44 
 
 8.4 
 
 9 
 
 40 
 
 
 15 
 
 1,320. 
 
 6,361.74 
 
 20.7 
 
 TABLE XXX. — Discharge from Corroded Pipes. 
 
 
 
 
 Approximate 
 
 Head in Feet. 
 
 Discharge per Minute. 
 
 
 Diam- 
 
 Ace 
 
 Amount of Cor- 
 
 
 
 
 
 No. 
 
 eter OF 
 
 OF 
 
 rosion PER LINEAR 
 
 
 
 Before 
 
 After 
 
 
 Pipe. 
 
 Pipe. 
 
 Yard. 
 
 Before 
 Cleaning. 
 
 After 
 Cleaning. 
 
 Cleaning. 
 
 Cleaning, 
 
 
 Inches. 
 
 Years. 
 
 Cu. Inches. 
 
 
 
 Imp. Gallons 
 
 Imp. Gallons 
 
 I 
 
 3 
 
 29 
 
 86.99 
 
 42 
 
 47 
 
 47 
 
 143 
 
 2 
 
 3 
 
 29 
 
 93- 
 
 54 
 
 56 
 
 79 
 
 188 
 
 3 
 
 3 
 
 29 
 
 93- 
 
 70 
 
 74 
 
 143 
 
 200 
 
 4 
 
 3 
 
 32 
 
 190. 
 
 77 
 
 82 
 
 16 f 
 
 150 
 
 5 
 
 3 
 
 32 
 
 190. 
 
 72 
 
 72 
 
 115 
 
 187 
 
 6 
 
 3 
 
 26 
 
 80. 
 
 56 
 
 62 
 
 35 
 
 220 
 
 7 
 
 3 
 
 26 
 
 88. 
 
 36 
 
 43 
 
 65 
 
 130 
 
 8 
 
 4 
 
 29 
 
 100. 
 
 40 
 
 45 
 
 69 
 
 "5 
 
 9 
 
 4 
 
 29 
 
 100. 
 
 38 
 
 42 
 
 125 
 
 107 
 
 These results are the average gaugings of five different trials taken once a week 
 on the same day. 
 
 On account of the action of most waters on cast-iron pipes it 
 is usual at the present time to protect the surface in some way 
 from corrosion. The process commonly employed is that devised 
 by Dr. R. Angus Smith : the newly cast pipes, which must be 
 free from rust, are heated to a temperature of some 500° Fahr., 
 and then dipped perpendicularly into a hot bath of coal-tar 
 pitch mixed with a small proportion of heavy coal oil. In. this 
 bath they are allowed to remain for a short time and then with- 
 drawn. The coating thus formed is firmly coherent and is unob- 
 
 * M. B. Jamieson : The internal corrosion of cast-iron pipes. Proc. Inst. Civ. 
 Eng. Gr. Br., Ixv (188 1), p 323. 
 
CONDUITS AND DISTRIBUTION PIPES. 
 
 209 
 
 jectionable from a sanitary point of view. It does not afford 
 absolute protection against rust, but it delays and diminishes the 
 action of the water to a great extent. The first line of pipes of 
 this description laid in this country was laid in Boston in 1858. 
 In 1876-77 some of these (20 in.) pipes were removed. "As 
 they were taken up their condition was observed. Their inner 
 surfaces were not entirely free from tuberculation, but were very 
 much more so than are the surfaces of uncoated pipes in this 
 city after they have been laid but a few years. The tubercles 
 were isolated, and were not in suiificient numbers or of suiificient 
 size to very materially interfere with the capacity of flow of the 
 pipes. They were very easily removed — more easily than from 
 uncoated pipes — seeming to have very little hold upon the tar 
 surface. 
 
 " Upon cleaning off the surface under a tubercle one would at 
 first suppose there had been simply a deposit, that no action had 
 been had either upon the iron or upon the coating ; but a more 
 careful examination would show that under the center of the 
 tubercle a portion of the iron, from the size of a pin head to that 
 of a small pea, had been transformed into a black substance that 
 could be easily cut with a knife, and had the appearance of plum- 
 bago. The inference drawn from the general appearance of the 
 pipes was that they would have lasted for an indefinite period." * 
 
 Some time since Professor Barff, of London, proposed to 
 protect articles of iron, among other things water pipes, from 
 corrosion, by covering them with an artificial coating of the black 
 oxide of iron. The coating is produced by exposing the 
 metal to superheated steam at a high temperature, and when 
 once formed it protects the iron from atmospheric and other 
 agencies which would corrode it. The process has been some- 
 what modified and is now known as the Bower-Barff process, and 
 promises to become a practical and valuable means of protecting 
 cast-iron pipes. Some wrought-iron pipes of this description 
 have been introduced as service pipes in Altona, and probably 
 elsewhere, but it is too soon for definite statements as to their 
 durability. 
 
 Cast-iron pipes are usually connected by the hub and spigot 
 joint ; the joints are first packed with tow or jute and then 
 
 * First Annual Report of the Boston Water Board. City Document No. 57. 
 Boston, 1877. 
 
 14 
 
2 lO WATER SUPPLY. 
 
 melted lead is run in and driven up firmly. It has been found 
 that the tow sometimes gives an unpleasant taste to the water, 
 and the Rivers Pollution Commission recommended that the 
 joints of the larger mains should be pointed up with Portland 
 cement on the inside to prevent the water from coming in con- 
 tact with the tow. 
 
 The sediment which accumulates in the pipes, especially in 
 low-lying districts, although it has generally a rusty appearance, 
 is not simply iron rust from the pipes. With the iron rust there 
 is always more or less earthy matter, and sometimes the sedi- 
 ment contains a large proportion of vegetable organisms, such 
 as the Crenothrix, already described. The following results were 
 obtained from the analysis of the rust in the Aberdeen pipes 
 already referred to : 
 
 Fron uncoated 4-in. Pipe From coated lo-in. Pipe 
 
 21 years in use. 15 years in use. 
 
 Organic and volatile matter 16.62 18.05 
 
 Sulphuric acid (SO;,) 0.60 1.08 
 
 Phosphoric acid Slight trace trace. ' 
 
 Magnetic oxide of iron 32.47 0-36 
 
 Peroxide of iron g.04 37.S5 
 
 Insoluble sandy matter 41-27 42.78 
 
 Lime trace o. 18 
 
 Wrought-iron pipes, coated with cement inside and out, have 
 been sometimes used, generally from motives of economy. They 
 are made by rolling up sheet iron and riveting the edges of the 
 sheet together as shown in Fig. 54. A longitudinal rib is some- 
 times employed as shown in Fig. 53. The lengths of pipes may 
 be telescoped together, but are usually connected by means of 
 an iron sleeve, filled in with cement. The durability of the 
 
 pipes is very dif- 
 ferent in different 
 localities, owing 
 mainly to dif- 
 ference in the 
 quality of the 
 workmanship. 
 Where the pipes 
 
 have been made 
 Fig. 53. Fig. 54. ^t, j- 
 
 under the direc- 
 tion of the water department, by day labor, they have sometimes 
 
CONDUITS AND DISTRIBUTION PIPES. 211 
 
 proved very satisfactory, but where the work is done by con- 
 tract it is difficult always to obtain the best results. 
 
 In the extreme West, wrought-iron pipes are used for con- 
 veying water for hydraulic operations and for purposes of water 
 supply, sometimes under a very great head. Water is conveyed 
 to Virginia City, Nevada, through such a main, the maximum 
 thickness of which is 0.34 inch, and which is exposed in some 
 parts of its length to a pressure of 1,800 feet head of water. The 
 pipes are coated with asphaltum to prevent rusting. 
 
 Wooden pipes, generally bored logs, have been used more or 
 less for conveying water, and are still employed in some sections 
 of the country. Detroit had at one time 130 miles of wooden 
 pipe, and i^ miles were laid in 1880. The logs used are sound 
 green tamarack, not less than 6 inches in diameter and 8 feet 
 long. The joints are covered with iron thimbles, and the pipes 
 last for 16 years or more and cost (or did cost) only about one- 
 fifth as much as iron. 
 
 Service pipes. — The service pipes for house distribution in 
 connection with a public water supply are generally of lead, this 
 metal being employed on account of the facility with which it 
 maybe worked. Lead pipes are also sometimes used for convey- 
 ing well or spring water to individual residences. Various waters 
 act very differently upon lead, some corroding it rapidly, others 
 only to a very slight extent, under similar circumstances. The 
 cause of the corrosion is to be sought in the dissolved oxygen, 
 of which all waters contain more or less, and in certain saline sub- 
 stances the presence of which determines a more violent action. 
 It is generally felt, for instance, that the presence of nitrates, 
 nitrites, and ammoniacal salts increases the action of water on 
 lead, while carbonates, sulphates, and notably phosphates, hinder 
 such action ; but while certain general statements may be truth- 
 fully made as the result of laboratory experiment and from the 
 analysis of waters whose action on lead has been learned by 
 experience, it is a rather hazardous thing for a chemist to pre- 
 dict, a priori, vjhdX v/ill be the effect of a particular water on lead 
 pipe under the conditions of ordinary practice. Next to no value 
 attaches to experiments made by immersing strips of sheet lead 
 in open or closed vessels containing the water under examma- 
 tion. In actual practice, many waters which would be pro- 
 nounced dangerous on the strength of such experiments, prove 
 
212 WATER SUPPLY. 
 
 entirely harmless. The pipes very soon become covered with a 
 naturally formed protective coating of difficultly soluble com- 
 pounds of lead ; and after a slight initial action, corrosion prac- 
 tically ceases if the pipes are kept constantly filled. 
 
 If it is felt necessary to make or to have made preliminary 
 laboratory experiments, they should be made by imitating as 
 nearly as possible the conditions of actual practice, and sufficient 
 time should be employed. The following extract from the report 
 on the examination of a soft surface water (where the time at 
 disposal was somewhat limited), will serve as an example : '" 
 
 "A coil of lOO feet of new one-quarter inch lead pipe was 
 taken and filled with the water under examination. The pipe 
 held 64 cubic inches of water, and the surface of lead was equal 
 to, about 900 square inches. The water was allowed to remain 
 in the pipe for 50 hours and then removed, a fresh supply being 
 introduced without allowing the air to come into contact with 
 the inside surface of the pipe. The water as drawn was quite 
 turbid, from the presence o-f the oxycarbonate of lead, and was 
 found to contain (the lead being calculated as metallic lead) : 
 
 Metallic Lead. 
 Parts per 100,000. Grains in U. S. gallon. 
 
 In solution 0.055 0.032 
 
 In suspension 1-257 O- 733 
 
 Total 1-317 0.765 
 
 " At the end of 70 hours the water in the pipe was drawn out 
 and; found to contain : 
 
 Metallic Lead- 
 Parts per 100,000, Grains in U. S. g^allon. 
 
 In solution o. 573 o. 334 
 
 In suspension 0.137 0.080 
 
 Total o. 710 0.414 
 
 "The pipe was then thoroughly washed from any loosely 
 adhering coating by allowing a rapid stream of Cochituate water 
 to flow through for some time. The Cochituate water was then 
 displaced by the water under examination, and this water was 
 allowed to remain in the pipe for 30 hours. This water, when 
 drawn, contained, both in solution and suspension, o. 157 part of 
 metallic lead in 100,000, or 0.092 grain to the U. S. gallon. The 
 
 * Report on the Waters of Flax Pond, made to the City Council of Chelsea, Mass.. 
 1875. 
 
SERVICE PIPES. 213 
 
 amount in suspension was not determined separately, as there 
 was no very considerable quantity visible to the eye. A fresh 
 portion was allowed to remain in the pipe for 40 hours, and then 
 contained in solution and suspension 0.184 part in 100,000, or 
 0.107 grain to the U. S. gallon." 
 
 At the conclusion of the experiment the action on the lead 
 had not ceased, even practically, but it had diminished very much, 
 and there was no doubt that, in practice, the water would act 
 very slightly on the pipes when in continuous use, as has proved 
 to be the case in Boston, New York, Glasgow, Manchester, and 
 many other places where the question has been discussed. 
 
 It may be said that, while with most waters the action on the 
 lead practically ceases, it probably never ceases absolutely. The 
 water of Lake Cochituate, as supplied in Boston, Mass., through 
 lead pipes, always contains traces of lead in solution. The 
 amount of lead taken up by the water in passing through some 
 150 feet of pipe which had been in use for some years, was found 
 to be only 0.03 part in 100,000, or less than 0.02 grain in the 
 U. S. gallon. Water which is allowed to remain in the pipe for 
 some time, or is drawn from the hot-water faucets, may contain 
 as much as o. i, or even 0.2 part in 100,000 (from 0.06 to o. 12 
 grain in the gallon), and wherever lead distribution pipes are in 
 use, it is safer always to run to waste enough water to clear the 
 pipes, and never to use, for drinking or for cooking, water which 
 has passed through the pipes while hot. A similar precaution 
 should be used in the case of new pipes : the water should be 
 wasted intermittently but freely for a number of days. There is 
 great difference in the susceptibility of different persons to lead 
 poisoning. It is thought that as little as one-fortieth of a grain to 
 the gallon has caused sickness, but one-tenth of a grain is usually 
 regarded as an outside limit. It is doubtful whether there are any 
 well authenticated cases of lead poisoning from the use of the 
 Cochituate water.* The Croton water supplied to New York 
 city is similar to the Boston water in its action on lead,"!" although 
 at least one case of poisoning has been reported, which was sup- 
 posed to be due to the daily use for some time of water which 
 had stood over night in the pipes. The practical experience in 
 
 * See Report of the Mass. State Board of Health, 1871. 
 
 t See Report of the Metropolitan Board of Health, New York, 1869, p. 420. 
 
214 WATER SUPPLY. 
 
 the use of lead pipe in the cities mentioned, and in many others, 
 shows that, as a rule, there is no danger in using lead pipes for 
 house distribution in connection with a public supply. 
 
 The most unfavorable situation for lead pipe is as suction 
 pipes in wells. Here the corrosion is often very rapid, and it is 
 rendered more violent by the fact that the continual changes of 
 level expose a longer or shorter portion of the pipe to the alter- 
 nate action of air and water. There are instances enough of 
 lead poisoning from this cause. 
 
 It may be remarked, in this connection, that the lead pipe 
 now in use, at least in the eastern part of the country, is much 
 inferior in strength and durability, and apparently more readily 
 corroded than that formerly in use. The lead now in the mar- 
 ket has been desilverized by the zinc process, and this seems to 
 give it a particular and disadvantageous character. 
 
 Other materials besides lead are used in the house service. 
 To block tin or to tin-lined lead pipes, if the latter are properly 
 made and properly put together, there is no objection on sani- 
 tary grounds. The corrosion of the tin by ordinary waters would 
 result in the formation of insoluble and harmless substances. As 
 to the suitability of the brass pipes which have been proposed, 
 there seems to be no exact information. To the various sorts 
 of " enameled " wrought-iron pipes which are in the market 
 there is no sanitary objection. The coating or enamel is gener- 
 ally some preparation of coal tar, with or without linseed oil, and 
 this sort of pipe is particularly adapted for use in wells, where a 
 portion' of the outer surface is exposed alternately to the action 
 of air and water ; unfortunately, the coating is not always per- 
 fect, and when the original surface of the pipe is exposed, rusting 
 begins. Zinced or " galvanized " iron, as it is called, is fully as bad 
 in respect to rusting. The pipes are prepared by dipping the 
 iron, previously well cleaned by means of dilute acid, into a bath 
 of melted zinc. The zinc adheres firmly to the surface of the 
 iron, and penetrates it to a certain extent, so that we do not deal 
 with a simple coating, such as we have on tinned iron, or on the 
 various forms of enameled pipe. The idea is that the zinc shall 
 protect the iron by virtue of a galvanic action between the 
 two metals, and it does protect the iron for a time. When the 
 pipes are exposed to the action of water, corrosion begins at 
 once : at first, the action is on the zinc alone, provided the origi- 
 
SERVICE PIPES. 215 
 
 nal iron was free from rust, and the treatment with zinc was 
 thorough ; but after a time the zinc which remains will cease to 
 protect the iron, and iron rust will begin to form. As regards 
 this action, it is simply a question of time. Water that has 
 passed through zinced pipes will be found almost always, if not 
 invariably, to contain zinc compounds, either in solution or in 
 suspension ; the amount, however, is generally very small. As 
 to the effect of such water on health, there is some difference of 
 opinion, but it is generally believed that the pipes may be safely 
 used.* 
 
 One of the best materials for service pipes is wrought iron 
 protected by the Bower-Barff process, provided practical experi- 
 ence justifies the theoretical expectations. To such pipes, coupled 
 without the use of red or white lead, there can be nothing supe- 
 rior from a sanitary point of view, and for use in wells and cis- 
 terns they will supply a very serious want. Ordinary wrought- 
 iron pipes, although possessing many advantages, have the great 
 disadvantage of rusting very readily : the iron rust is harmless 
 but unsightly in drinking water, and may render the water unfit 
 for culinary purposes and for use in the laundry. 
 
 * For a full discussion of this subject, see Dr. Boardman's paper in the Report of 
 the Mass. State Board of Health for 1874. 
 
BIBLIOGRAPHY. 
 
 The following list of books makes no claim to being exhaustive. It includes 
 most of the works referred to in the preceding pages, and may perhaps be 
 described as a list of such works as would together make a fair nucleus for a 
 librai-y of water supply, other than from a strictly engineering standpoint. 
 Papers in periodical publications are not included. 
 
 I. — Works of a General Character, mainly from an Engineering 
 
 Point of View. 
 
 Fanning, J. T. : A Practical Treatise on Water-Supply Engineering, etc. 
 8vo, pp. 650. Van Nostrand, New York, 1877. 
 
 Grahn, E. : Die stadtische Wasserversorgung. 3 vols. Svo. Vol. I. Mi'in- 
 chen, 1877. [Contains Statistik der stadtischen Wasserversorgung : Beschrei- 
 bung der Anlagen in Bau und Betrieb.] 
 
 Humber, William : A Cotnprehensive Treatise on the Water Supply ot 
 Cities and Towns, etc. Folio, pp. 378, and many plates. London, Crosby, 
 Lockwood & Co., 1877. [An American edition was published by Geo. H. 
 Frost, Chicago.] 
 
 n. — Works of a General Character, mainly from a Chemical or 
 Sanitary Point of View. 
 
 Buck : Hygiene and Public Health. 2 vols. Svo. New York, Wm. Wood 
 & Co., 1879. 
 
 Denton, J. Bailey : Sanitary Engineering. 8vo, pp. 429, with many plates. 
 Spon, London,- 1877. 
 
 Fischer : Die chemische- Technologic des Wassers. 8vo. Braunschweig, 
 1878-80. 
 
 Fodor : Boden und Wasser. Braunschweig, 1882. 
 
 Great Britain : Report of the Royal Commission on Water-Supply, with 
 Minutes of Evidence. Parliamentary Documents. 4to. London, 1869-70. 
 
 Great Britain : Sixth Report of the Commissioners appomted in 1868 to in- 
 quire into the Best Means of Preventing the Pollution of Rivers. Domestic 
 Water Supply of Great Britain. 4to. London, 1876. 
 
 [This report contains, besides complete statistics of the water supply of 
 Great Britain, considerations and accounts of experiments on the following 
 topics : 
 
 On the alleged self-purification of polluted rivers. 
 
 On the propagation of epidemics by potable water. 
 
BIBLIOGRAPHY. 2X7 
 
 On the alleged influence of the hardness of water upon health. 
 
 On the superiority of soft over hard water in cooking. 
 
 On the softening of hard water. 
 
 On the improvement of potable water by filtration. 
 
 On the deterioration of potable water by transmission through mains and 
 service pipes. 
 
 On the constant and intermittent systems of water supply.] 
 
 Lefort, Jules : Traits de Chimie Hydrologique. 8vo, pp. 798. Deuxi&me 
 edition. Paris, 1873. 
 
 Lersch, Dr. B. M. : Hydrochemie oder Handbuch der Chemie der natur- 
 lichen Wasser, 2 Auflage. 8vo, pp. 718. Bonn, 1870. 
 
 Parkes, E. A., M.D.: Manual of Practical Hygiene. Fifth edition. 8vo. 
 London and Philadelphia, 1878. 
 
 Reichardt: Grundlagen zur Beurtheilung des Trinkwassers. 3te Auflage. 
 8vo, pp. 107. Jena, 1875. 
 
 Sander, Dr. Friedrich : Handbuch der offentlichen Gesundheitspflege. 
 Bvo, pp. 503. Leipzig, 1877. 
 
 Wolffhiigel : Wasserversorgung. [Aus dam Handbuch der Hygiene und 
 der Gewerbekrankheiten, von Pettenkofer und Ziemssen. ] 8vo, pp. 244. 
 Leipzig, 1882. 
 
 IIL — On the Pollution of Streams. 
 
 Great Britain : Rivers Pollution Commission, appointed in 1865. Three 
 Reports. Parliamentary Documents. 4to. London, 1866-67. 
 
 Great Britain : Rivers Pollution Commission, appointed in 1868. Si.x Re- 
 ports. Parliamentary Documents. 4to. London, 1870-74. 
 
 Massachusetts : Seventh Annual Report of the State Board of Health, 
 containing a special report on the pollution of streams, etc. Bvo. Boston, 
 1876. 
 
 Paris : Assainissement de la Seine. Epuration et Utilisation des Eaux 
 d'Egout. Documents administratifs ; Enqu^te ; Annexes. 3 vols. 8vo, with 
 plates. Paris, 1876. 
 
 IV. — On Filtration, Ground Water, Wells, etc. 
 
 Belgrand, M. : Les Travaux souterrains de Paris. Etudes prgliminaires. 
 La Seine. Regime de la Pluie, des Sources, des Eaux courantes. Applica- 
 tions k 1' Agriculture. 8vo, pp. 612, with atlas. Paris, Dunod, 1875. [Espe- 
 cially chap, vii, Des nappes d'eau souterrains, and chap, xxvi, Du filtrage 
 des eaux, etc.] 
 
 Berlin : Vorarbeiten zu einer zukiinftigen Wasserversorgung der Stadt 
 Berlin. Ausgefuhrt in den Jahren 1868 und 1869 von L. A. Veitmeyer. 
 Svo, pp. 368, with atlas. Berlin, Reimer, 1871. [Especially pp. 109-130 — 
 experiments on the ground water in the neighborhood of the Tegeler See.] 
 
 Berlin : Reinigung und Entwasserung Berlins. Berichte iiber mehrere 
 auf Veranlassung des Magistrats der konigl. Haupt-und Residenzstadt Berlin 
 Versuche und Untersuchungen. 12 Hefte, with 3 Anhange. Svo. Berlin, 
 
2l8 WATER SUPPLY. 
 
 Hirscljwald, 1870-76. [Especially Heft v, " iiber die Grundwasserverhalt- 
 nisse," with a great number of profiles.] 
 
 Brooklyn, K. Y. : The Brooklyn Water Works and Sewers. Prepared 
 and printed by order of the Board of Water Commissioners. 4to, pp. 159, 
 with 59 lith. plates. New York, Van No.strand, 1 867. 
 
 Darcy, Henry : Les fontaines publiques de la ville de Dijon. 4to, pp. 647, 
 with atlas. [Especially Appendix D, Filtrage.] 
 
 Dresden : Das Wasserwerk der Stadt Dresden erbaut in den Jahren 1871- 
 1874, von B. Salbach. 8vo. In three parts, with atlases containing many 
 plates. Halle, Knapp, 1874-76. 
 
 Dupasquier, A. : Des Eaux de Source et des Eaux de Riviere comparees, 
 etc. i2mo, pp. 414. Paris et Lyon, 1840. 
 
 Dupuit, J. : Traits de la Conduite et de la Distribution des Eaux, etc. 
 Suivi de la Description des Filtres naturelles de Toulouse p.ar D'Aubisson. 
 4to, with atlas. Paris, 1854. 
 
 Fischer, Dr. F. : Das Trinkwasser, seine Besch.affenheit, Untersuchung 
 und Reinigung unter Beriicksichtigung der Brunnenwasser Hannover. 8vo, 
 pp. 63. Hannover, 1873. 
 
 Gottisheim: Das unterirdische Basel. 8vo, pp. 72. Basel, 1868. [2d 
 Edition, 1873.] 
 
 Grahn and Meyer : Reisebericht einer von Hamburg nach Paris und Lon- 
 don ausgesandten Commission iiber kiinstliche centrale Sandfiltration zur 
 Wasserversorgung von Stadten, und iiber Filtration in kleinen Massstabe. 
 Von E. Grahn und F. Andreas Meyer. 8vo, pp. 153. Hamburg, Meissner, 
 1877. [Especially Anlage 3, "Historische Notizen iiber kiinstliche Filtration 
 im kleineren Massstabe."] 
 
 Grimaud, de C'ux : Des Eaux publiques, etc. 8vo, pp. 348. Paris, Dezo- 
 bry, 1863. [Especially chap, xiv, " de la clarification des eaux publiques."] 
 
 Halle: Das Wasserwerk der Stadt Halle, erbaut in den Jahren 1867 und 
 1868. Von B. Salbach. Folio, with atlas. Halle, Knapp, 1871. 
 
 Kirkwood, J. P. : Report on the Filtration of River Waters for the Supply 
 of Cities, as practised in Europe. 410, pp. 178, with 30 plates. New York, 
 Van Nostrand, 1869. 
 
 Munich : Berichte iiber die Verhandlungen und die Arbeiten der Commis- 
 sion fiir Wasserversorgung, Canalisation, und Abfuhr. Erster Bericht, 1874- 
 75 ; Zweiter Bericht, 1876, und Anhiinge, 1877. 4to, with plans and profiles. 
 Mvinchen, Miihlth.aler, 1876, 1877. 
 
 Nichols, W. R. : On. the Filtration of Potable Water. Reprinted from the 
 Ninth Annual Report of the Mass. State Board of Health. 8vo, pp. 93. New 
 York, Van Nostrand, 1879. 
 
 Pielke ; Mittheilungeii iiber naturliche und kiinstliche Sandfiltration, nach 
 Betriebsresultaten der Beriiner Wasserwerke vor dem Stralauer Thor. 8vo, 
 pp. 75. Beriin, 1881. 
 
 Schorer, Th. : Lubeck's Trinkwasser, 8vo, pp. 284. Lubeck, Seelig, 1877. 
 [Especially pp. 248-257, describing the deterioration of water by vegetable 
 growth and decay, etc.] 
 
BIBLIOGRAPHY. 2IQ 
 
 Spon, Ernest : The present Practice of Sinking and Boring- Wells, izmo, 
 pp. 216. London, Spon, 1875. 
 
 Ward, F. O. : IVEoyens de cr^er des sources artificielles d'eau pure pour 
 Bruxelles. 8vo, pp. 106. Bruxelles, Decq, 1853. 
 
 Wiebel, Dr. F, : Die Fluss- und Bodenvvasser Hamburgs. Chemische 
 Beitrage zur Analyse gewohnlicher Lauf- Nutz- und Trinkwiisser sowie zu 
 iler Frage der Wasserversorgung grosser Stiidte von sanitaren und gewerb- 
 lichem Standpunkte, 4to, pp. 152. Hamburg, Meissner, 1876. 
 
 Wolff; Der Untergrund und das Trinkwasser der Stadte. ate Auflao-e. 
 8vo, pp. 60. Erfurt, 1873. 
 
 V. — On the uSanitary Examination of Water. 
 
 Eyferth, B. : Die mikroscopischen Siisswasserbewohner in gedrangter 
 Uebersicht. 8vo, pp. 60. Braunschweig, 1877. 
 
 Fox, Cornelius B., M.D. : Sanitary Examination of Water, Air and Food. 
 8vo, pp. 508. London, 1878. 
 
 Frankland, Dr. E. : Water Analysis for Sanitary Purposes. i2mo, pp. 139. 
 London, Van Voorst, 1880. 
 
 Hassall : Microscopical Examination of Water Supplied to the Inhabitants 
 of London, etc. London, 1851. 
 
 Kubcl, Dr. Wilhelm : Anleitung zur Untersuchung von Wasser, u. s. w. 
 8vo, pp. 184. Zweite Auflage von Dr. Ferd. Tiemann. Braunschweig, 1874. 
 
 Macdonald, J. D., M.D. : A Guide to the Microscopical Examination of 
 Drinking Water. 8vo, pp. 65 and 24 plates. London and Philadelphia, 1875. 
 
 Neuville: Des Eaux de Paris. Essai d'Analyse micrographique compare'e. 
 4to, pp. 63. 15 plates. Paris, 1880. 
 
 Schiitzenberger: On Fermentations. i2mo, pp. 331. Intern. Sci. Series. 
 New York, 1876. [This contains a description of Schiitzenberger's method of 
 determining dissolved oxygen, pp. 108 and foil.] 
 
 Wanklyn, J. A. : Water Analysis: a Practical Treatise on the Examination 
 of Potable Water. By J. A. W. and E. T. Chapman. i2mo, pp. 182. Fifth 
 edition, re-written by J. A. Wanklyn. London, 1879. 
 
 VI. — Miscellaneous. 
 
 Boston, Mass. ; Report of Cochituate Water Board, 1874. [Contains are- 
 print of a report by Geo. f. Deacon, Borough Engineer, Liverpool, Eng., on 
 the subject of " Waste."] 
 
 Boston, Mass.: Report on Waste of Water' (May 25, 1882). Boston City 
 Document No. 78, 1882. 
 
 Bowditch: Public Hygiene in America. 8vo. Boston, 1877. 
 
 Deacon: The Constant Supply and Waste of Water. A paper read before 
 the Society of Arts, May 19, 1882. 410, London, 1882. 
 
 Magnin: The Bacteria, translated by Dr. Geo. H. Sternberg, U. S. A. 
 8vo, pp. 227. Bo.ston, 1880. 
 
 Nageli: Die niederen Pilze. Munchen, 1877. 
 
220 WATER SUPPLY. 
 
 Nageli: Untersuchungen iiber niedere Pilze. 8vd, pp. 285. Munchen 
 und Leipzig, 1882. [Tiiis contains Buchner's paper referred to on page 21.] 
 
 De Ranee : The Water Supply of England and Wales. 8vo, pp. 623. Lon- 
 don, 1882. 
 
 Kuessner und Pott: Die acuten Infectionskrankheiten. 8vo, pp. 460. 
 Braunschweig, 1882. 
 
 Parry: Water, its Composition, Collection, and Distribution. i2mo, pp. 
 184. London, 1881. 
 
 Rowan: Boiler Incrustation and Corrosion. i8mo, pp. 48. New York, 
 Van Nostrand, 1876. 
 
 Wilson : Treatise on Steam Boilers. Third edition. i2mo, pp. 328. Lon- 
 don, 1875. 
 
TABLES FOR CALCULATION. 
 
 221 
 
 TABLE XXXI.— For the Conversion of Degrees of Fahrenheit's Scale into those of 
 the Centigrade Thermometer. 
 
 Fahr. 
 
 Cent. 
 
 Fahr. 
 
 Cent. 
 
 Fahr. 
 
 Cent. 
 
 Fahr. 
 
 Cent. 
 
 
 
 - 17-7 
 
 54 
 
 12.2 
 
 107 
 
 41.6 
 
 160 
 
 71. 1 
 
 I 
 
 - 17.2 
 
 55 
 
 13.7 
 
 108 
 
 42.2 
 
 j6i 
 
 71.6 
 
 2 
 
 — 16.6 
 
 5& 
 
 13-3 
 
 log 
 
 42.7 
 
 162 
 
 72.2 
 
 3 
 
 — 16. 1 
 
 57 
 
 13-8 
 
 no 
 
 43-3 
 
 163 
 
 72.7 
 
 4 
 
 - 15-5 
 
 58 
 
 14.4 
 
 III 
 
 43.8 
 
 164 
 
 73-3 
 
 5 
 
 — 15 
 
 59 
 
 15 
 
 . 112 
 
 44-4 
 
 165 
 
 73-8 
 
 6 
 
 — 14.4 
 
 60 
 
 15.5 
 
 113 
 
 45 
 
 166 
 
 74-4 
 
 7 
 
 - 13.8 
 
 61 
 
 16. 1 
 
 H4 
 
 45-5 
 
 167 
 
 75 
 
 8 
 
 - 13-3 
 
 62 
 
 16.6 
 
 "5 
 
 46.1 
 
 168 
 
 75-5 
 
 9 
 
 - 12.7 
 
 63 
 
 17.2 
 
 ij6 
 
 46.6 
 
 169 
 
 76.1 
 
 10 
 
 — 12.2 
 
 04 
 
 17-7 
 
 117 
 
 47.2 
 
 170 
 
 76.6 
 
 II 
 
 — II. 6 
 
 65 
 
 18.3 
 
 118 
 
 47-7 
 
 171 
 
 77.2 
 
 12 
 
 — ii:i 
 
 65 
 
 18. S 
 
 119 
 
 48.3 
 
 172 
 
 77-7 
 
 13 
 
 - 10,5 
 
 67 
 
 19.4 
 
 1 20 
 
 48.8 
 
 173 
 
 78.3 
 
 14 
 
 — 10 
 
 68 
 
 20 
 
 121 
 
 49-4 
 
 174 
 
 73.8 
 
 15 
 
 - 9.4 
 
 6g 
 
 20.5 
 
 122 
 
 50 
 
 175 
 
 79-4 
 
 i6 
 
 — 8.8 
 
 70 
 
 21. 1 
 
 123 
 
 50.5 
 
 176 
 
 80 
 
 17 
 
 - 8.3 
 
 71 
 
 21 .6 
 
 124 
 
 5I-I 
 
 177 
 
 80.5 
 
 i8 
 
 - 7-7 
 
 72 
 
 22.2 
 
 125 
 
 51.6 
 
 178 
 
 81. 1 
 
 19 
 
 - 7-2 
 
 73 
 
 22.7 
 
 126 
 
 52.2 
 
 179 
 
 81.6 
 
 20 
 
 - 6.6 
 
 74 
 
 23-3 
 
 127 
 
 52.7 
 
 180 
 
 82.2 
 
 21 
 
 - 6.1 
 
 75 
 
 23.8 
 
 128 
 
 53-3 
 
 181 
 
 82.7 
 
 22 
 
 - 5-5 
 
 76 
 
 24.4 
 
 129 
 
 53-8 
 
 182 
 
 83-3 
 
 23 
 
 - 5 
 
 77 
 
 25 
 
 130 
 
 54-4 ■ 
 
 183 
 
 83.8 
 
 24 
 
 - 4.4 
 
 78 
 
 25-5 
 
 "3C 
 
 55 
 
 184 
 
 84.4 
 
 25 
 
 - 3-8 
 
 79 
 
 26,1 
 
 132 
 
 55-5 
 
 185 
 
 85 
 
 26 
 
 - 3-3 
 
 80 
 
 26.6 
 
 133 
 
 56.1 
 
 186 
 
 85.5 
 
 27 
 
 - 2.7 
 
 81 
 
 27.2 
 
 134 
 
 56.6 
 
 187 
 
 86.1 
 
 28 
 
 — 2.2 
 
 82 
 
 27.7 
 
 135 
 
 57-2 
 
 18S 
 
 86.6 
 
 2g 
 
 - 1.6 
 
 83 
 
 28.3 
 
 136 
 
 57-7 
 
 189 
 
 87.2 
 
 30 
 
 — I.I 
 
 84 
 
 2S.8 
 
 137 
 
 58. 3 
 
 190 
 
 87.7 
 
 31 
 
 - 0.5 
 
 85 
 
 29.4 
 
 138 
 
 58.8 
 
 191 
 
 88.3 
 
 32 
 
 
 
 86 
 
 30 
 
 139 
 
 59-4 
 
 192 
 
 88.8 
 
 33 
 
 0-5 
 
 87 
 
 30.5 
 
 140 
 
 6o 
 
 193 
 
 89.4 
 
 34 
 
 I.I 
 
 88 
 
 31. 1 
 
 141 
 
 60.5 
 
 194 
 
 90 
 
 35 
 
 1.6 
 
 89 
 
 31.6 
 
 142 
 
 61. 1 
 
 195 
 
 90.5 
 
 36 
 
 2.2 
 
 90 
 
 32.2 
 
 143 
 
 61.6 
 
 196 
 
 91. 1 
 
 37 
 
 2.7 
 
 gt 
 
 32.7 
 
 144 
 
 ' 62 . 2 
 
 197 
 
 gi.6 
 
 38 
 
 3-3 
 
 93 
 
 33-3 
 
 145 
 
 62.7 
 
 198 
 
 92.2 
 
 39 
 
 3-8 
 
 93 
 
 33-8 
 
 146 
 
 63.3 
 
 199 
 
 92.7 
 
 40 
 
 4-4 
 
 94 
 
 34-4 
 
 147 
 
 63.8 
 
 200 
 
 93-3 
 
 41 
 
 5 
 
 95 
 
 35 
 
 148 
 
 64.4 
 
 201 
 
 93.8 
 
 42 
 
 5-5 
 
 96 
 
 35-5 
 
 149 
 
 65 
 
 202 
 
 94-4 
 
 43- 
 
 6.1 
 
 97 
 
 36.1 
 
 150 
 
 65.5 
 
 203 
 
 95 
 
 44 
 
 6.6 
 
 93 
 
 36.6 
 
 151 
 
 66.1 
 
 204 
 
 95-5 
 
 45 
 
 7-2 
 
 99 
 
 37-2 
 
 152 
 
 66.6 
 
 205 
 
 96.1 
 
 46 
 
 7-7 
 
 ■ ICO 
 
 37-7 
 
 153 
 
 67.2 
 
 206 
 
 96.6 
 
 47 
 
 8.3 
 
 lOI 
 
 38-3 
 
 154 
 
 67.7 
 
 207 
 
 97.2 
 
 48 
 
 8.8 
 
 102 
 
 38.8 
 
 155 
 
 68.3 
 
 208 
 
 97-7 
 
 49 
 
 9.4 
 
 103 
 
 39-4 
 
 156 
 
 68.8 
 
 209 
 
 98-3 
 
 50 
 
 10 
 
 104 ' 
 
 40 
 
 157 
 
 69.4 
 
 210 
 
 98.8 
 
 51 
 
 10.5 
 
 105 
 
 40.5 
 
 158 
 
 70 
 
 211 
 
 99-4 
 
 52 
 
 II. I 
 
 106 
 
 41. 1 
 
 159 
 
 70.5 
 
 212 
 
 100. 
 
 53 
 
 II. 6 
 
 
 
 
 
 
 
222 
 
 WATER SUPPLY. 
 
 TABLE XXXII. — For the Conversion of Degrees of the Cent^sde Thermometer into 
 Degrees of Fahrenheit's Scale, 
 
 Cent. 
 
 Fahr. 
 
 Cent. 
 
 Fahr. 
 
 Cent. 
 
 Fahk. 
 
 Cent. 
 
 Fahr. 
 
 O 
 
 32.0 
 
 26 
 
 78.8 
 
 51 
 
 123.8 
 
 76 
 
 168.8 
 
 I 
 
 33.8 
 
 27 
 
 80.6 
 
 52 
 
 125.6 
 
 77 
 
 170.6 
 
 2 
 
 35-6 
 
 28 
 
 82.4 
 
 53 
 
 127.4 
 
 78 
 
 172.4 
 
 3 
 
 37-4 
 
 29 
 
 84.2 
 
 54 
 
 129.2 
 
 79 
 
 174.2 
 
 4 
 
 39-2 
 
 30 
 
 86.0 
 
 55 
 
 131-0 
 
 80 
 
 176.0 
 
 5 
 
 41.0 
 
 31 
 
 87.8 , 
 
 5" 
 
 132.8 
 
 8i 
 
 177-8 
 
 6 
 
 42.8 
 
 32 
 
 89.6 
 
 57 
 
 134-6 
 
 82 
 
 179.6 
 
 7 
 
 44.6 
 
 33 
 
 91.4 
 
 53 
 
 136.4 
 
 83 
 
 181.4 
 
 8 
 
 46.4 
 
 34 
 
 93-2 
 
 59 
 
 138.2 
 
 84 
 
 183.2 
 
 9 
 
 48.2 
 
 35 
 
 950 
 
 60 
 
 140.0 
 
 85 
 
 1S5.0 
 
 10 
 
 50.0 
 
 36 
 
 96. 8 
 
 61 
 
 141. 8 
 
 86 
 
 186.8 
 
 II 
 
 51.8 
 
 37 
 
 9S.6 
 
 62 
 
 143.6 
 
 87 
 
 ^188.6 
 
 12 
 
 53.6 
 
 38 
 
 100.4 
 
 63 
 
 145-4 
 
 88 
 
 190.4 
 
 ^3 . 
 
 55-4 
 
 39 
 
 102.2 
 
 64 
 
 147.2 
 
 Sg 
 
 192.2 
 
 14 
 
 57-2 
 
 40 
 
 104.0 
 
 65 
 
 149.0 
 
 go 
 
 194.0 
 
 15 
 
 59.0 
 
 41 
 
 105.8 
 
 66 
 
 150.8 
 
 91 
 
 195-8 
 
 l6 
 
 608 
 
 42 
 
 107.6 
 
 67 
 
 152.6 
 
 92 
 
 197.6 
 
 17 
 
 62.6 
 
 43 
 
 109.4 
 
 68 
 
 154-4 
 
 93 
 
 199.4 
 
 i8 
 
 64.4 
 
 44 
 
 III. 2 
 
 69 
 
 156-2 
 
 94 
 
 201.2 
 
 19 
 
 66.2 
 
 45 
 
 113. 
 
 70 
 
 158.0 
 
 95 
 
 203.0 
 
 20 
 
 68.0 
 
 46 
 
 114. 8 
 
 71 
 
 159.8 
 
 96 
 
 204.8 
 
 21 
 
 69.8 
 
 47 
 
 T16.6 
 
 72 
 
 161.6 
 
 97 
 
 206.6 
 
 22 
 
 71.6 
 
 48 
 
 118. 4 
 
 73 
 
 163.4 
 
 98 
 
 208.4 
 
 23 
 
 73-4 
 
 49 
 
 120.2 
 
 74 
 
 165.2 
 
 99 
 
 210.2 
 
 24 
 
 75.2 
 
 50 
 
 122.0 
 
 75 
 
 167.0 
 
 100 
 
 212.0 
 
 25 
 
 77.0 
 
 
 
 
 
 
 
 Metric System of Weights and Measures. 
 Weiirhts, 
 
 1 Milligram mgm. 
 
 I Centigram 
 
 1 Decigram 
 
 I Gram grm. 
 
 = o.ooi gram. 
 
 = o.ci " 
 
 = 0.1 " 
 
 = 1.0 " 
 
 I Gram 
 1 Dekagram 
 1 Hectogram 
 T Kilogram 
 
 gnu. 
 
 kilo. 
 
 =: I gram. 
 = lo grams. 
 = 100 " 
 = 1000 " 
 
 I Millimeter m.m. 
 
 I Centimeter cm. 
 
 I Decimeter d.m. 
 
 1 Meter m. 
 
 = O.OOI 
 
 = o.oi 
 = 0.1 
 = i.o 
 
 Measures of Length 
 meter. 
 
 1 Meter 
 
 I Dekameter 
 
 1 Hectometer 
 
 1 Kilometer kilom. 
 
 = I meter. 
 = lo meters. 
 = 100 " 
 = 1000 " 
 
 Measures of Volume. 
 
 I Cubic Centimeter 
 1 Cubic Decimeter 
 I Cubic Meter 
 
 c.c. or cm. 
 
 dm. 
 
 m.' 
 
 = O.OOI liter. 
 = 1.000 " 
 = 1000.000 " 
 
WEIGHTS AND MEASURES. 
 
 223 
 
 s 
 
 o 
 
 > 
 
 < 
 > 
 
 h 
 Z 
 
 a 
 < 
 
 3 
 
 w 
 
 I 
 
 X 
 X 
 X 
 
 .w 
 ij 
 m 
 < 
 
 0. 
 
 on 
 
 
 1-^ o* 
 
 
 CO 
 
 
 
 
 
 
 
 -a- or^o 
 
 gS 
 
 
 
 o 
 
 r- c^ c^ 
 
 c^ 
 
 
 M 
 
 c<^ t^ 
 
 <-J;S 
 
 
 
 
 
 M -i- 
 
 •a 
 
 
 
 
 
 
 S 
 
 
 N 
 
 
 t 
 
 
 CO r^ 
 
 fig 
 
 r- iri en 
 
 tn Tj- i-l CO 
 Q O O N 
 
 0000 
 
 vO ■^^ ro 
 ro O u^ 
 cOvO tn 
 
 M H C 
 
 ' 6 O 6 y 
 
 en ^ M CO O 
 
 en 
 
 c 
 
 
 
 
 l-l 
 
 C>0 H 
 
 
 
 
 6 
 
 
 M 'O 
 
 cn 
 
 M en c^ 
 
 
 CO 
 
 « N 
 
 
 
 
 ■c 
 
 c w -o 
 
 
 
 CI 
 
 a 
 
 
 CI 
 
 H 
 
 
 
 
 
 
 o 
 
 C^ 00 u^ 1"^ 
 
 Cn Hi HI U-) 
 
 O -i- O vo 
 O vo CO 00 
 M N Tf Hi 
 
 HI HI o ^^ "^ 
 
 ^ fc/3 
 
 o <u o 
 
 pU '^i H-l 
 
 _ CJ o o 
 
 ^ (i ttj *^ IS iS 
 . p: .ti 3 :3 3 
 
 
 
 ^ 
 
 in 
 
 "rt 
 
 
 [/) 
 
 b 
 
 
 
 ^ 
 
 Cil 
 
 
 
 
 
 
 
 r 
 
 
 -a 
 
 e 
 
 en 
 en 
 
 
 
 
 
 
 vn 
 
 
 
 
 en 
 
 Uj 
 
 
 c^ 
 
 
 
 
 
 
 r 
 
 
 r1 
 
 c 
 
 
 "o 
 
 0: 
 
 T3 
 
 s 
 
 
 
 l 
 
 
 -O 
 
 
 fe 
 
 
 
 
 ^_ 
 
 rt 
 
 C 
 
 ^^ 
 
 2 
 
 '•xi 
 
 3 
 
 
 s 
 
 
 
 ^ 
 
 rC 
 
 r/i 
 
 S 
 
 is 
 
 H 
 
 X 
 
 w 
 
 
 
 
 
 
 
 
 
 )-• 
 
 ^ 
 
 'Tl 
 
 
 >> 
 
 
 
 
 
 
 
 
 !> 
 
 rt 
 
 
 
 en 
 
 ^ 
 
 1 
 
 3 
 
 
 
 
 bJ3 
 
 c 
 W 
 
 H en 
 «co" 
 
 'a 
 
 
 "■ .5 H S ^ s 
 
 c -^ 2 ■" a "s 
 
 is " S 8 g^ g 
 
 1;^ Ca U] >< •- >-, 'w 
 
 >. O Q, 2 W *S *^ 
 
 -^ 4J ^ ^ rSH (2 r^ 
 
 . T5 U L^ ■*-• Qj S 
 
 . >■ C u 
 a. c H /3 
 
 
 « 3 
 
 g 1: 
 
 ^ ^ ^ " 
 
 <U {i o 
 
 •a 
 
 c 
 
 C 3 
 en o 
 
 
 
 3 3 rt 
 
 O +J -.-t tn 
 " 3 01 -3 
 
 o .3 
 
 a. .. 
 S - 
 
 ?■ bjj u '^ 
 
 f f^ ^ CO tS ^ 
 '^ o TOM ^*^ iX 
 
 3 
 
 bJ3 *J ^ 
 
 tn 
 
 ho d x: 
 
 ^ o ? 
 .b '^ rS 
 
 b* O " 
 
 
 tn 
 
 u 
 
 ^ ^ rt 
 
 C (N g S ^ 
 
 •- " 2 -o 8 
 
 -ess 
 
 C<J 
 
 iffl 
 
 w 
 
 i-l 00 
 
 w 
 w 
 
 _r PL, a. 
 
 in (n ^ 
 
 - CO O 
 
 .5 m 
 
 3 en pj 
 
 ' % be 
 
 •<il 
 
 e) 
 
 en 
 
 W 
 
 u 
 
 — -a u o) ^ jj 
 2 II ^ .S -^ 
 
 ^ U 0) [fl '^ 
 
 f2 a s x5 'o 
 
 •S u - "So a 
 
 H H g W g 
 « -^ 2i « o 
 
 •3 
 
 .2 o .S 
 
 •S 3 o 
 
 2 " — 
 
 60 Si ° 
 
 .S « £ 
 
 o 
 
 — U C 
 
 2 5 a .2 5. 
 
 <U 
 
224 
 
 WATER SUPPLY. 
 
 TABLE XXXIV. — For Facililating Approximate Cakulaiions from one 
 Denojnination into another. 
 
 Feet. 
 
 > 
 Meters. 
 
 Pounds 
 Avoirdu- 
 pois. 
 
 Kilo- 
 grams. 
 
 U.S. 
 Gallons. 
 
 Imperial 
 Gallons. 
 
 Liters. 
 
 Cubic 
 Meters. 
 
 r 
 
 0.30 
 
 I 
 
 0.45 
 
 I 
 
 0.83 
 
 3-79 
 
 0.0038 
 
 2 
 
 0.61 
 
 2 
 
 0.91 
 
 n 
 
 1.67 
 
 7.57 
 
 O.C076 
 
 3 
 
 0.91 
 
 3 
 
 1.36 
 
 3 
 
 2.50 
 
 11.36 
 
 0.0114 
 
 4 
 
 1.22 
 
 4 
 
 1. 81 
 
 4 
 
 3-33 
 
 15.14 
 
 0.0151 
 
 5 
 
 1.52 
 
 5 
 
 2.26 
 
 5 
 
 4.17 
 
 18.93 
 
 0.0189 
 
 6 
 
 1.83 
 
 6 
 
 2.72 
 
 6 
 
 5.00 
 
 22.71 
 
 0.0227 
 
 7 
 
 2.13 
 
 7 
 
 3.18 
 
 7 
 
 5.83 
 
 26.50 
 
 0.0265 
 
 8 
 
 2.44 
 
 8 
 
 3.62 
 
 8 
 
 6.66 
 
 30.38 
 
 0.0303 
 
 9 
 
 2.74 
 
 9 
 
 4.08 
 
 9 
 
 7.50 
 
 34.07 
 
 0.0341 
 
 3.28 
 
 I 
 
 2.20 
 
 I 
 
 1.20 
 
 I 
 
 4-54 
 
 0.0045 
 
 6.56 
 
 2 
 
 4.41 
 
 2" 
 
 2.40 
 
 2 
 
 908 
 
 0.0091 
 
 9.84 
 
 3 
 
 6.61 
 
 3 
 
 3- 60 
 
 3 
 
 13-63 
 
 0.0136 
 
 13.12 
 
 4 
 
 8.82 
 
 4 
 
 4.80 
 
 4 
 
 18.17 
 
 0.0182 
 
 16.40 
 
 5 
 
 11.02 
 
 5 
 
 6.00 
 
 5 
 
 22.71 
 
 0.0227 
 
 ig.69 
 
 6 
 
 13.23 
 
 6 
 
 7.20 
 
 6 
 
 27.26 
 
 0.0273 
 
 22.97 
 
 7 
 
 15-43 
 
 7 
 
 8.40 
 
 7 
 
 31.80 
 
 0.0318 
 
 26.25 
 
 8 
 
 17.64 
 
 8 
 
 9.60 
 
 8 
 
 36.34 
 
 0.0363 
 
 29.53 
 
 9 
 
 19.84 
 
 9 
 
 10.80 
 
 9 
 
 40.89 
 
 0.0409 
 
 Miles. 
 
 Kilo- 
 
 Kilo- 
 
 Miles. 
 
 
 meters. 
 
 meters. 
 
 
 0.26 
 
 0.22 
 
 I 
 
 O.OOI 
 
 I 
 
 I.61 
 
 I 
 
 0.62 
 
 2 
 
 3-22 
 
 2 
 
 1.24 
 
 0.53 
 
 0.44 
 
 2 
 
 0.002 
 
 3 
 
 4-83 
 
 3 
 
 1.86 
 
 0.79 
 
 0.66 
 
 3 
 
 0.003 
 
 4 
 
 6.44 
 
 4 
 
 2.49 
 
 1.05 
 
 0.88 
 
 4 
 
 0.004 
 
 5 
 
 8.05 
 
 5 
 
 3." 
 
 1.32 
 
 1. 10 
 
 5 
 
 0.005 
 
 6 
 
 9.66 
 
 6 
 
 3.73 
 
 1. 58 
 
 1.32 
 
 6 
 
 0.006 
 
 7 
 
 11.27 
 
 7 
 
 4-35 
 
 1.85 
 
 1-54 
 
 7 
 
 0.007 
 
 8 
 
 12.87 
 
 8 
 
 4-97 
 
 2. II 
 
 1.76 
 
 8 
 
 0.008 
 
 9 
 
 14.48 
 
 9 
 
 5 59. 
 
 2.3S 
 
 1.98 
 
 9 
 
 0.009 
 
TABLES FOR CALCULATION. 
 
 225 
 
 TABLE XXXV. — For the Conversion of Paris per 100,000 into Grains per Gallon, 
 a7id vice versa : also, for Comparing Degrees of Hardness. 
 
 
 
 
 Degrees of Hardness. || 
 
 
 Grains in 
 
 Grains in 
 
 
 English : 
 
 Grains 
 
 CaCOa in 
 
 
 Parts in 100,000. 
 
 U. S. Standard 
 
 Imperial 
 
 French : 
 
 German : 
 
 
 Gallon. 
 
 Gallon. 
 
 Parts CaCO 3 
 
 Parts CaO 
 
 
 
 
 in ICO, 000. 
 
 imp. Gallon. 
 
 in ico,ooo. 
 
 I 
 
 0.5837 
 
 0.7 
 
 I 
 
 0.7 
 
 0.6 
 
 2 
 
 I. 1674 
 
 1.4 
 
 2 
 
 1-4 
 
 I.I 
 
 3 
 
 I. 7512 
 
 2.1 
 
 3 
 
 2.1 
 
 1-7 
 
 ' 4 
 
 2.3349 
 
 2.9 
 
 4 
 
 2.8 
 
 2.2 
 
 5 
 
 2.9186 
 
 3-5 
 
 5 
 
 3-5 
 
 2.8 
 
 6 
 
 3-5023 
 
 4-2 
 
 6 
 
 4.2 
 
 3.4 
 
 7 
 
 4.0861 
 
 4.9 
 
 7 
 
 4-9 
 
 3-9 
 
 8 
 
 4.6698 
 
 5-6 
 
 8 
 
 5.6 
 
 4.5 i 
 
 9 
 
 5-2535 
 
 6.3 
 
 9 
 
 6.3 
 
 5-0 
 
 1.7131 
 
 1 
 
 I. 1992 
 
 1-4 
 
 1 
 
 0.8 
 
 3.4262 
 
 2 
 
 2.3983 
 
 2-9 
 
 2 
 
 1.6 
 
 ■ 5-1393 
 
 3 
 
 3-5975 
 
 4-3 
 
 3 
 
 2-4 
 
 6.8524 
 
 4 
 
 4.7967 
 
 6.7 
 
 4 
 
 3.2 
 
 8.5655 
 
 5 
 
 5-9958 
 
 7-1 
 
 5 
 
 4.0 
 
 10.2786 
 
 6 
 
 7.1950 
 
 8.6 
 
 6 
 
 4.8 
 
 11.9917 
 
 7 
 
 8.3942 
 
 10. 
 
 7 
 
 5-6 
 
 13.7048 
 
 8 
 
 9-5934 
 
 ri-4 
 
 8 
 
 6.4 
 
 15.4179 
 
 9 
 
 10.7925 
 
 12.8 
 
 9 
 
 7-2 
 
 1.4286 
 
 0.8339 
 
 I 
 
 1.8 
 
 1.2 
 
 I 
 
 2.8571 
 
 1.6678 
 
 2 
 
 3.6 
 
 2.5 
 
 2 
 
 4.2857 
 
 2.5017 
 
 3 
 
 5.4 
 
 3-8 
 
 3 
 
 5 • 7143 
 
 3-3356 
 
 4 
 
 7-1 
 
 5 
 
 4 
 
 7.1428 
 
 4.1695 
 
 5 
 
 9-0 
 
 6-3 
 
 5 
 
 8.5714 
 
 5 0033 
 
 6 
 
 10.7 
 
 7-5 
 
 6 
 
 10.0000 
 
 5-8372 
 
 7 
 
 12.6 
 
 8.8 
 
 7 
 
 11.4286 
 
 6.6711 
 
 8 
 
 14-3 
 
 10. 
 
 8 
 
 12.8571 
 
 7.5050 
 
 9 
 
 16. 1 
 
 11-3 
 
 9 
 
 15 
 
INDEX. 
 
 Aberdeen, corrosion of pipes in, 208. 
 
 Absorplion of gases, 11. 
 
 Aeration, 150. 
 
 Africa, artesian wells in, 134. 
 
 Air, composition of, 12. 
 
 Aire and Calder rivers, pollution of, 58. 
 
 Alabama, artesian wells in, 135. 
 
 Albuminoid ammonia, 39. 
 
 Algae at Berlin, 157. 
 
 in deep-seated water, 143. 
 in water supplies, 85. 
 spores not removed by sand filtra- 
 tion, 165. 
 Algeria, artesian wells in, 135. 
 Altenburg, consumption of water in, 194. 
 
 water supply of, 207. 
 Alton, III., filters at, 168. 
 Altona, consumption of water in, 194. 
 filtration at, 156, 
 use of Barff's process in, 209. 
 Alum, treatment of water with, 187. 
 Ammonia, coefficient of absorption of, II. 
 
 determination of, 37. 
 Ammonia method of water analysis, 37, 
 
 38. 
 Amsterdam, water supply, 121. 
 Anabaena circinalis, 87, 88. 
 Analyses, methods of reporting, 8, 30. 
 Analysis, methods of, 29. 
 Animal life, 79. 
 Anthrax, 20. 
 
 Aquamalarial fever, 52, 100. 
 Artesian wells at Crenelle, Paris, 137. 
 St. Louis, 137. 
 Charleston, S. C, 135. 
 in Alabama, 135. 
 Algeria, 134. 
 China, 134. 
 examination of (Table), 
 
 144. 
 pollution of, 143. 
 uncertainty of, 137. 
 Artificial ice, 53. 
 Atkins's cistern filters, 178. 
 
 Bacillus, -li, 20. 
 
 anthracis, 20. 
 
 subtilis, 21. 
 Bacteria, 19, 45, 67. 
 
 removed by spongy iron, 173. 
 
 Baird's system of distillation, 190. 
 Baltimore, Md., water supply, 204. 
 Bamberg, consumption of water in; 194. 
 Bangor, Me., filters at, 169. 
 Barff's process, 209, 215. 
 Belgium, pollution of streams in, 72. 
 Belgrand, quoted, 120^ 
 Bell's waste detector, 201. 
 Berlin, experience with meters, 204. 
 filtration at, 157. 
 
 ground water measurement at, no. 
 water supply, trouble in, 125. 
 Bibliography, 216. 
 Biological examination of water, 45. 
 Birmingham, consumption of water in, 
 
 194. 
 Bischof's experiments on bacteria, 173. 
 
 spongy iron for filters, 166, 327. 
 Blackburn, Eng,, consumption of water 
 
 in, 194. 
 Blackstone River, pollution of, 59, 62. 
 Bohemia, spring and well waters in, 145. 
 Boiler scale, 181. 
 
 Bone coal for filters, 170, 334, 336. 
 Bonn, consumption of water in, 194. 
 Boston, consumption of water in, 195. 
 corrosion of iron pipes in, 209. 
 lead pipes in, 213. 
 waste water inspection, 201. 
 water, examination of, loi, 104. 
 Bower-Barff process. See Barff. 
 Brackett, Dexter, quoted, 194, 195, 200, 
 
 201, 205. 
 Bradford Beck, pollution of, 58. 
 Braunschweig, consumption of water in, 
 
 194. 
 Breslau, experience with meters, 204. 
 Brewer, Dr., quoted, 52. 
 Brick for filters, 168, 332. 
 Brooklyn, daily consumption in, 195. 
 
 inspection of plumbers' fittings, 
 
 205. 
 water supply of, 105. 
 well in Prospect Park, in. 
 Buchanan, quoted, 15. 
 Buchner's experiments on Bacillus an- 
 thracis, 21. 
 
 Cambridge, Mass., consumption of water 
 in, 195. 
 
2 28 
 
 INDEX. 
 
 Carbide of iron for filters, i66. 
 Carbonate of lime, solubility of, 9. 
 Carbonic acid, coefficient of absorption, 
 II. 
 solution of, 12. 
 ■ in natural waters, 31. 
 Cast-iron pipes, corrosion of, 207. 
 
 method of coupling, 209. 
 Cellulose used for filters, 180. 
 Chandler, analysis of Croton water, 8. 
 Charbon, 20. 
 
 Cliarleston, S. C, artesian wells at, 135. 
 Chemical analysis, value of, 44. 
 solution, 2. 
 
 treatment of water, 187. 
 with alum, 187. 
 with lime, 183, 188. 
 with perchloride of iron, 187. 
 with permanganate of pot- 
 ash, 188. 
 Chicago, consumption of water in, 195. 
 
 River, pollution of, 59. 
 China, artesian wells in, 134. 
 Chlorides, significance of, 33, 68. 
 Chlorine, a means of tracing pollution, 
 131. 
 in natural waters, 33. 
 estimation of, 33. 
 Cholera and drinking water, 22. 
 Church's waste indicator, 379. 
 Cincinnati, consumption of water in, 195 . 
 
 detection of waste, 201 . 
 Cistern filters, 176. 
 
 water, examination of, 50. 
 
 (Table), 51. 
 Cisterns, sediment in, 50. 
 Clark's process, 183. 
 Classification of waters, 17, 42, 43, 
 Clathrocystis aeruginosa, 88. 
 Clinton, Iowa, filters at, 168. 
 Clyde, pollution of, 59. 
 Coelosphaerium, 87. 
 Cologne (Coin), consumption of water in, 
 
 194. 
 Color, method for estimating, 149. 
 Conduits, 206. 
 
 growth of sponge in, 207. 
 various, length of, 207. 
 Consumption of water in American cities, 
 195. 
 European cities, 
 194. 
 Corroded pipes, diminished flow through, 
 
 207. 
 Corrosion of cast-iron pipes, 207. 
 Couste, quoted, 5. 
 Crenothrix Kilhniana, 126, 210. 
 Crookes, Prof., quoted, 188. 
 Croton water, action on lead, 213. 
 
 analysis of, 8. 
 Cucumber taste, 91. 
 Cyclops quadricornis, 70. 
 
 Danube, suspended matter in, 57. 
 Danzig, water supply of, 207. 
 Daphnia pulex, 79. 
 
 Davis, J. P. , cost of sand filtration, 165. 
 Deacon's system in Liverpool, 199. 
 Glasgow, 200. 
 Boston, 201 . 
 Deacon's wastewater meter, 196-199. 
 Deep-seated water, alga^ in, 143. 
 
 as a source of supply, 
 
 133-145- 
 characteristics of, 142. 
 examination of, 142. 
 (Tables), 144, 145. 
 pollution of, 143. 
 Deep sea thermometer, 95. 
 Deep wells, 134, 139. 
 
 examination of (Table), 144. 
 Degrees of hardness, 34. 
 
 different thermometers com- 
 pared, 221, 222. 
 Deposition as means of self-purification, 
 
 67. 
 De Ranee, quoted, 137, 141. 
 Desmids, 86. 
 Detroit, consumption of water in, 195. 
 
 wooden pipes in, 211. 
 Diatoms, 86. 
 
 Dilution as means of natural purification, 
 68. 
 a guaranty of safety, 71. 
 Discharge from corroded pipes, 208. 
 Disease and drinking water, 17-28. 
 Dissolved solids, effect of, 10. 
 
 estimation of, 32. 
 Distillation, 189. 
 
 Baird's system, 190. 
 Normandy's system, 192. 
 Distribution pipes, 206. 
 Double supply, 206. 
 Dresden, consumption of water in, 194. 
 
 ground water measurements, no. 
 water supply of, to8, 120. 
 Drinking water and disease, 17-28. 
 best, 133. 
 theory, 22. 
 Drinking, water most suitable for, 27. 
 Driven wells, 109, 11 2- 11 7. 
 
 principle of, 114. 
 Dublin, consumption of water in, 194. ' 
 Dubuque, Iowa, water supply of, 138. 
 Durance, suspended matter in, 57. 
 
 Edinburgh, consumption of water in, 194. 
 Elbe filtered at Altona, 156. 
 
 pollution of, 71. 
 Emmerich, Dr., quoted, 24, 26, 45. 
 England, pollution of streams in, 73. 
 
 Fall River, Mass., consumption of water 
 in, 195. 
 use of meters in, 203. 
 
INDEX. 
 
 22g 
 
 Ferrous sulphate test. 35. 
 Filter, Piefke's, 180. 
 
 the multifold, 179. 
 Filter basin at Waltham, Mass., iig. 
 Filter beds, advantage of covering, 163. 
 construction of, 151, 155. 
 frequency of cleaning, 1 57. 
 method of cleaning, 152. 
 Filter gallery at Columbus, O., 108. 
 Halle, 108. 
 Lowell, 107. 
 Taunton, 108. 
 Filters, household, bone coal for, 170, 172. 
 for cisterns, 176-178. 
 requirements of, 170. 
 reversible, 171. 
 silicated carbon for, 172. 
 simple form of, 174. 
 spongy iron for, 173. 
 wood charcoal for, 1 74. 
 Filter press. Porter's, 185. 
 Filtration, 151. 
 
 at Alton, 111., 168. 
 Altona, 156. 
 Antwerp, 166. 
 Bangor, Me., i6g. 
 Berlin, 157, 158, 160, 164. 
 Clinton, Iowa, 168. 
 Hudson, N. Y., 161, 163. 
 London, 159, 160. 
 Magdeburg, 164. 
 Malone, N. Y., 168. 
 Marshalltown, Iowa, 168. 
 Poughkeepsie, N. Y , 162. 
 Wakefield, Eng., 166. 
 Zurich, 157. 
 extent of, 151. 
 expense of, 165. 
 for manufactories, 179. 
 household, i6g. 
 
 materials used, 169. 
 in winter, 164. 
 natural, 106. 117. 
 principles of, 152. 
 rate of, 152, 156, 158. 
 (See also Filters, household.) 
 Flint, Dr. Austin, quoted, 23. 
 Flushing, water supply of, 193. 
 Forel, Dr., quoted, 80. 
 Fox, Dr. C. B., quoted, 44. 
 .France, pollution of streams in, 59, 73. 
 Frankfort, consumption of water in, 194. 
 temperature of water, 207. 
 water supply, 207. 
 Frankland, Dr., quoted, 35, 40, 41. 
 Frankland's classification of water, 41. 
 method of analysis, 37, 39. 
 report on Antwerp filters, 166. 
 
 Galvanized iron, corrosion of, 214. 
 Ganges, suspended matter in, 57. 
 Gases, coefficient of absorption, 11. 
 
 Gases in natural waters, 31. 
 solution of, 10. 
 
 supersaturated solutions of, 12. 
 Germ theory, 18. 
 
 Germany, pollution of streams in, 59. 
 Glasgow, consumption of water in, 194. 
 
 inspection of fittings in, 205. 
 
 lead pipe in, 213. 
 
 water supply of, 207. 
 
 waste water detection, 200. 
 Glasgow water, examination of, loi. 
 Gold, solubility of, 15. 
 Gotha, water supply of, 207. 
 Grahn, quoted, 151. 
 Great Britain, pollution of streams in, 58, 
 
 73- 
 Greaves, Mr. Chas., on filter sand, 155. 
 Grenelle, Paris, artesian well at, 137. 
 Ground water as source of supply, 105, 
 132. 
 defined, 105. 
 effect of pumping on, 109- 
 
 112. 
 examination of, 122. 
 
 (Table), 124. 
 hardness of, 120. 
 inclination of, 105. 
 temperature of, 118. 
 utilization of, 106, 109. 
 Ground-water supphes, difficulties of, 123, 
 128. 
 at Berlin, 125-127. 
 Halle, 127. 
 Leipzig, 123. 
 Lille, 127. 
 
 Halle, water supply of, 108. 
 
 trouble with, 127. 
 Hamburg, consumption of water in, 194. 
 Hannover, consumption of water in, 194. 
 Hardness, degrees of, 34. 
 
 determination of, 33. 
 
 permanent, 33, 1 86. 
 
 temporary, 34, 181. 
 Hard water, 17, 181. 
 
 softening of, 181-1S7. 
 wholesomeness of, 18. 
 Hatton, experiments on spongy iron, 173. 
 Head, term defined, 156. 
 Heisch's test, 46. 
 Hoadley, J. C, quoted, 116. 
 Hooghly, character of water, 16. 
 Household filtration (See Filtration). 
 Hudson, N. Y., filtration at, 161, 163. 
 Hull, Eng., consumption of water at, 194. 
 Humus, 84. 
 Hunt, T. Sterry, quoted, 16. 
 
 Ice, chemical examination of, 54, 55. 
 
 impure, 52. , 
 
 I in filtered water, 179. 
 
 I natural and artificial, 5a 
 
230 
 
 INDEX. 
 
 Impounding reservoirs, 85. 
 Improvement of natural water, 146. 
 by aeration, 150. 
 chemical processes, 187. 
 Clarlc's process, 181. 
 distillation, 189. 
 filtration, 151. 
 sedimentation, 147. 
 softening, 181. 
 storage, 149. 
 Infusoria, 83. 
 Iron, carbide of, 166. 
 
 compounds of, in ground water, 123, 
 
 125, 129. 
 perchloride of, 187. 
 pipes, corrosion of, 207. 
 protection of, 208. 
 sediment in, 210. 
 spongy, 166. 
 Irrawaddy, suspended matter in, 57. 
 Irwell, pollution of, 58. 
 
 Jamieson, M. B., quoted, 207. 
 
 Karlsri;he, consumption of water in, 
 
 194. 
 Kassel, consumption of water in, 194. 
 Kerner, quoted, 207. 
 Kirkwood, quoted, 155, 164. 
 Koch, quoted, 20. 
 Koch's biological method, 46. 
 
 Latham, quoted, 92, 141, 143. 
 
 Latham's apparatus for tempering water, 
 
 93- 
 Lawrence, Mass., manufacturing indus- 
 tries at, 63. 
 Laws against poUutioh of water supplies, 
 
 71-78. 
 Lead pipes, action of water on, 211-214. 
 Lead poisoning, 213. 
 Leeds, Eng., consumption of water in, 
 
 194. 
 Leipzig, consumption of water in, 194. 
 
 water supply of, 123. 
 Leyden, water supply of, 121. 
 Lille, trouble with water supply, 127. 
 Lime, water treated with, 183, 188. 
 Liverpool, Eng. , consumption of water in, 
 194. 
 inspection of fittings in, 205 . 
 softening of water at, 185. 
 water supply, 138. 
 London, results of filtration at, 159, 160. 
 temperature of water supplies, 
 
 93- 
 water, examination of, 102. 
 water supply, 102, 138. 
 wells in and near, 137. 
 Long Island, N. Y., ground water on, 105, 
 
 III, 121. 
 Loss on ignition, 37. 
 
 Lowell, Mass., consumption of water in, 
 
 195- 
 filtering gallery at. 107. 
 manufacturing refuse at, 
 
 63- 
 
 Maas, suspended matter in, 57. 
 Magdeburg, experience with meters, 204. 
 
 filter beds at, 164. 
 Manchester, Eng., consumption of water 
 in, 194. 
 
 inspection of fittings 
 in, 205. 
 
 lead pipe in, 213. 
 Mallet, Prof., quoted, 24, 25, 35, 40,42. 
 Malone, N. Y., filters at, l68. 
 Marshalltown, Iowa, filters at, 168. 
 Memphis, cisterns in, 50. 
 Merrimack river, pollution of, 62, 63. 
 Meters for checking waste, 203. 
 
 German experience with, 204. 
 Microbes, 19. 
 Micrococcus, ,-ci, 20. 
 Microzymes, 19. 
 Milzbrand, 20. 
 Mississippi filtered at Alton, 168. 
 
 sediment in, 16, 57. 
 Mixed solutions, condition of, 7. 
 Mountain fever, 52, 100. 
 Multifold filter, 179. 
 Munich, water supply of, 207. 
 
 Natural filtration, 106, 117. 
 Nageli, quoted, 19. 
 Nessler test, delicacy of, 34. 
 New Orleans, cisterns in, 49. 
 
 wells in, 130. 
 New River water, filtered, 160. 
 New York, inspection of fittings, 205. 
 lead pipes in, 213. 
 water supply, 194, 207. 
 wells in, 138. 
 Nitrites and nitrates, 35. 
 
 significance of, 36. 
 Nitrogen, coefficient of absorption, 1 1 . 
 (combined) in water, 34. 
 in natural waters, 31. 
 organic, 34. 
 Nodularia, 89. 
 
 Normandy's system of distillation, 192. 
 Nostocs, 87. 
 
 Odors and tastes of surf ace waters, 80, 84, 
 
 85,90. 
 Ohio River, pollution of, 70. 
 Organic and volatile matter, 37. 
 Organic carbon, 40. 
 
 matter, 24, 36, 38. 
 
 nitrogen, 34. 
 Oxidation, a means of purification, 66. 
 Oxygen, coefficient of absorption, II. 
 
 in natural waters, 31. 
 
INDEX. 
 
 231 
 
 Paris, consumption of water in, 194. 
 
 water supply of, 207. 
 Passaic River, pollution of, 59, 170. 
 Pawtucket, R. I., use of meters in, 203. 
 Permanganate method of analysis, 37. 
 
 of potash, a purifying 
 agent, i88. 
 in analysis, 37. 
 Philadelphia, consumption of water in, 
 
 195. 
 Physical solution, i. 
 Piefkc, quoted, 157, 161. . 
 Piefke's filter, 180. 
 Pipes, brass, 214. 
 
 enameled, 214. 
 galvanized iron, 214. 
 iron {Sde Iron pipes), 
 lead, 211. 
 service, 211. 
 tin-lined lead, 2:4. 
 wooden, 211. 
 wrought iron, 210. 
 Plants, aquatic, S3. 
 Polluting liquids defined, 74. 
 Pollution of streams, 57, 78. 
 
 prevention of, 71—78. 
 of wells, 12S. 
 Popular tests, 46. 
 Porter-Clark process, i S4. 
 Po, suspended matter in, 57. 
 Potamogeton, 83. 
 
 Potash, permanganate of, 37, 188. 
 Poughkeepsie, algx at, 90. 
 
 filtration at, 161, 162. 
 Previous sewage contamination, 35. 
 Protection of iron pipes, 208. 
 Providence, consumption of water in, 195. 
 inspection of fittings, 205. 
 use of meters in, 203. 
 Prussia, pollution of streams in, 72. 
 Purification of water by freezing, 53. 
 {See also Improvement.) 
 
 Quantity and waste, 194, 206. 
 
 necessary for public supply, 
 194. 
 
 Rain water, analyses of. 48. 
 
 as source of supply, 48-55. 
 
 requires filtration, 175. 
 
 storage of, 49. 
 Remsen's investigations at Boston, 80, 91. 
 Rhine, suspended matter in, 57. 
 Rhone, suspended matter in, 57. 
 River Lea water filtered, 159, 160. 
 Rivers Pollution Commission, quoted, 41, 
 48, 66, 74, 102, 103, I44> 145, 159. 160, 
 166. 
 Rivers Pollution Prevention act (1876), 74. 
 
 Salbach's photometer, 148. 
 Samples, collection of, 47. 
 
 Sand filtration, 151. 
 
 character of sand, 154. 
 details of practice, 154. 
 expense of, 165. 
 in the U. S., 161. 
 principles of, 152. 
 results of, 159. 
 Saturated solutions, 6. 
 Schizomycetes, ig. 
 Schiitzenberger's oxygen method, 31. 
 Schuylkill River, pollution of, 59. 
 Sediment in iron pipes, 210. 
 river waters, 57. 
 Sedimentation, 147. 
 Seine, pollution of, 59, 61. 
 Self-purifying power of streams, 63. 
 Service pipes, 2 E 1-2 15. 
 Settling basins, 146, 147. 
 Sharpies, quoted, I20. 
 Sheffield, consumption of water in, 194. 
 Silicated carbon for filters, 172. 
 Smart, Dr., quoted, 49, 52, 99, 100, 130. 
 
 simple form of filter, 174. 
 Smith's process for coating pipes, 208. 
 Soda-water, I2, 13. 
 Softening of hard water, 18 c. 
 Solids in solution, estimation of, 32. 
 
 solution of, I. 
 Solubility, cilrves of, 4. 
 
 of ammonia, ti. 
 carbonic acid, 11. 
 gases, 10, II. 
 liquids, 13. 
 metallic gold, 15. 
 nitrogen, 11. 
 solids, I. 
 
 sulphate of lime, 5. 
 Solution and suspension distinguished, 14. 
 Solution, I. 
 
 chemical, 2. 
 
 means of hastening, 9, 13. 
 of gases, 10. 
 liquids, 13. 
 solids, I. 
 physical, I. 
 Solutions, mixed, 7. 
 
 saturated, 6. 
 supersaturated, 6. 
 Solvent, term defined, 3. 
 Spaltpilze, ig." 
 
 Spencer's carbide of iron, 159, 166. 
 Spha;rozyga, 87. 
 Spirillum, -la, 20. 
 Spithead, England, wells at, 121. 
 Splenic fever, cause of, 20. 
 Sponge for filters, 168. 
 fresh-water, 80. 
 growth of in conduits, 207. 
 Spongilla fluviatilis, 80. 
 
 analysis of. 82. 
 figui-e of, Si. 
 spicules of, 81, 82. 
 
232 
 
 INDEX. 
 
 Spongy iron for filters, l66, 173. 
 S))ree filtered at Berlin, 157, 160. 
 Springlield, Mass., algx at, go. 
 Springs, 133. 
 
 effect of barometric pressure on, 
 141. 
 Standards of purity, 40. 
 St. Louis, artesian well at, 137. 
 
 consumption of water at, 195. 
 settling basins at, 147. 
 waste of water at, 195. 
 Storage, effect of, 149. 
 Storer, Prof., quoted, 6. 
 Streams, pollution of, 57. 
 
 self -purification of, 63. 
 turbidity of, 56. 
 
 table, 57. 
 Sulphate of iron test, 35. 
 
 lime, solubility of, 5. 
 Sulphuretted hydrogen in ponded water, 
 
 85. 
 Supersaturated solutions, 6, 12. 
 Surface water, animal and vegetable life 
 in, 79-90. 
 as a source of supply, 56. 
 examination of, 96. 
 
 (Tables), 103, 104. 
 tastes and odors of, 90. 
 temperature-of, 91, 
 variations of, 100. 
 Suspended matter, determination of, 29. 
 in Missouri R., 147. 
 in streams, 57. 
 Suspension defined, 14. 
 
 Tables for facilitating calculations, 221- 
 
 225. 
 Tank filters, 172. 
 Taunton, Mass., filter gallery at, 108. 
 
 ground water near, 105. 
 Temperature, change of in conduits, 207. 
 means of measuring, 95. 
 of Fresh Pond, Mass., 94. 
 ground water supplies, 
 
 n8. 
 London water, 93. 
 surface waters, 91. 
 Thames water filtered, 159, 160. 
 The Hague, water supply of, 121. 
 
 Thermometers, 95. 
 
 Tidy, Dr., classification of waters, 42. 
 Toulouse, filter gallery at, 119. 
 Turbidity, estimation of, 148. 
 
 of streams. 56. 
 Turbid water, clarification of, 15. 
 Typhoid fever and drinking water, 22. 
 
 UlxiK, quoted, 71. 
 
 Unpolluted water from various sources 
 (Table), 145. 
 
 Vegetable growth on filter beds, 164. 
 Vibrio (-nes), 20. 
 
 Vienna, water supply of, 133, 207. 
 Virginia City, Hcv., conduit, 211. 
 Vistula, suspended matter in, 57. 
 Vlissingen. See Flushing. 
 
 Wanklvn, quoted, 41. 
 Wanklyn's method of analysis, 38. 
 Waste of water, 194-206. 
 
 detection of, Bell's system, 201. 
 
 Church's method, 202. 
 Deacon's system, 196. 
 
 in winter, 205. 
 
 prevention of, 196. 
 Water analysis, 2g. 
 Waters, classification of, 17. 
 Weights and measures, tables of, 221- 
 
 225. 
 Well at Whiston, Eng., 139, 140. 
 
 Brooklyn, N. Y., III. 
 Wells, artesian, 133. 
 
 deep, 134, 139. 
 
 effect of pumping from, log. 
 
 near salt water, 121. 
 
 pollution of, 128. 
 
 shallow, 106. 
 Well water, examination of, 131. 
 
 (Table), 132. 
 Whiston, Eng., deep well at, 139, 140. 
 Whitma.., T. J., quoted, igs. 
 Wood charcoal for filters, 168, 174. 
 Wooden pipes, 211. 
 Worcester, sewage of, 62. 
 Wrought-iron pipes, 2IO. 
 
 Zurich, filtration at, 157. 
 Zymotic diseases, 18.