n a~) DI QIQ ^;, n - _ C3 S. ~:- "i"`:i~ i':e:a::1~r,.:c/.,s.:I I~~ri;;,,~.~"i 91;~~ ~'ed:~:: -:~:;, ~~ '~ a E ~~; -r n r ~. ~,;~ i~ (n '' ~~~ ~~ t':~ M tr " ",t~ C, ~' I) a ~~ o Z d~..:bd~ rid.' ' B Cf . ~ ',.1:'~ '';~crc ~~ ~ ~~'... -,~;::,,..-~.~~ B..i ...~ zi.-; ':~4 '=.~; "'\'' " U~i::.t '' a' r0* ' ' "' ~~~*i;~r~.:iv~~ —i;Y iF- Y -IYm -aarr~~IE-Y-~;-l:.Ir: —:~::~~i; ~;;~~i~ r.i~ c r9 C. 7 A., pI 11 I~ WASTE OF WATER IN NEW YORK AND ITS REDUCTION BY METERS AND INSPECTION A I WASTE OF WATER IN NEW YORK AND ITS REDUCTION BY METERS AND INSPECTION A REPORT BY JAMES H. FUERTES, C. E., TO THE COMMITTEE ON WATER-SUPPLY OF THE MERCHANTS' ASSOCIATION OF NEW YORK ALSO A DIGEST OF LAWS GOVERNING THE USE OF WATER-METERS IN NEW YORK, BY ALFRED L. MARILLEY AND A SUMMARY OF PRESENT CONDITIONS RELATING TO THE WATER-SUPPLY OF NEW YORK, BY J. HAMPDEN DOUGHERTY JUNE, I906 THE MERCHANTS' ASSOCIATION OF NEW YORK COPYRIGHT, 1900 BY THE MERCHANTS' AsSOCIATION OF NEW YORK COMMITTEE ON WATER SUPPLY HENRY R. TOWNE, Chairmna EDWARD P. NORTH, C. E. J. HAMPDEN DOUGHERTY HENRY W. GOODRICH E. E. OLCOTT ERNST J. LEDERLE, PH. D. FREDERICK B. DE BERARD, Secretary 149971 I TABLE OF CONTENTS PAGE PREFATORY: REPORT OF COMMITTEE ON WATER SUPPLY. 1 WASTE OF WATER IN NEW YORK AND ITS REDUCTION BY METERS AND INSPECTION By James H. Fuertes, C. E.. 11 Summary of Conclusions........ 13 Part I: Methods for Controlling Loss of Water 17 Part II: The Unaccounted-for Water.. 44 Part III: Non-Domestic Use of Water.... 50 Part IV: Water Required for Domestic and Public Uses.......... 54 Part V: Leakage and Wastage...... 60 Part VI: On the Amount of Water that May Be Had from the Croton Water-Shed. 65 Part VII: Future Conditions... 72 Part VIII: Practicable Methods of Controlling Leakage and Wastage in New York City............ 81 Part IX: The Introduction of Meters in New York as a Business Proposition.. 91 APPENDIX A: UNACCOUNTED-FOR WATER... 107 APPENDIX B: WASTE OF METERED WATER.... 113 APPENDIX C: LEAKAGE FROM STREET MAINS... 117 APPENDIX D: WASTE REDUCTION IN AMERICAN AND GERMAN CITIES...... 127 Part I: American Cities... 127 Part II: German Cities.... 220 PAGE Group I: Cities in Which Waste Reduction Has Accomplished by Metering....... 220 Group II: Cities in Which but a Small Proportion of the Supply Is Metered...... 227 Group III: Cities in Which a Large Proportion of the Supply Is Metered..... 229 APPENDIX E: LIST OF REFERENCES.... 231 DIGEST OF LAWS GOVERNING THE USE OF WATER METERS IN NEW YORK CITY: By Alfred L. Marilley...... 245 APPENDIX A: WATER RATES AND REGULATIONS.. 251 SUMMARY OF PRESENT CONDITIONS RELATING TO THE WATER SUPPLY OF THE CITY OF NEW YORK: By Hon. J. Hampden Dougherty, Former Commissioner of Water Supply, Gas and Electricity....... 253 INDEX.................. 269 PREFATORY REPORT OF COMMITTEE ON WAT E R S UPPLY REPORT OF COMMITTEE ON WATER SUPPLY. To the Board of Directors of the Merchants' Association of New York GENTLEMEN: One of the most important problems in connection with the City's water supply is the prevention of waste and leakage. The plans recently adopted by the Board of Water Supply, and approved by the State Water Commission, assure an ample supply of water after the expiration of eight or ten years. In the interim the danger of a water famine will be constant. In dry seasons the flow of the Croton River is frequently less than 50,000,000 gallons per day, while the present daily consumption of water in Manhattan and The Bronx is about 300,000,000 gallons. The City is therefore dependent upon stored water to supply its needs during the periods when the flow of the river is deficient. The existing storage is insufficient to carry the City over a protracted dry season, unless the present rate of consumption be materially reduced. In 1899 and 1900 the City was dependent for 255 days upon its stored water. At the present rate of consumption, in case of continued drought the stored supply would be exhausted in 245 days. Owing to growth of population and other causes, the daily consumption of water is increasing at the rate of 15,000,000 gallons each year. It will probably be at least eight years before any material increase of supply can be obtained. Before the new supply is available, therefore, the draft upon the storage during a dry season will have so increased as to exhaust it in about 175 days. The consumption of water cannot safely exceed 290,000,000 gallons per day when the flow of the Croton River is less than the daily consumption; otherwise the stored water may be exhausted before the end of a drought, and the City thus be left almost or entirely without water. 3 Prefatory It is certain that consumption must be curtailed in an increasing degree during the dry season of the next few years, as a precaution against the complete exhaustion of the water supply in case of protracted drought. A restricted allowance of water at times, with much inconvenience and even privation, is almost inevitable. The alternative is a probable water famine. It is therefore of the first importance that the available supply be conserved by the elimination of waste and leakage, so far as that elimination is practicable. It has been widely assumed that extreme waste is responsible for the large per capita consumption of water in this City. The actual extent of the waste is undetermined, the data being incomplete and not sufficient to sustain the contention, frequently made, that the major portion of all the water coming into the City is lost by leakage. The statements which assert such extreme leakage are not based upon ascertained facts, are largely conjectural, and are not worthy of serious consideration in an examination of this subject. In every city having a public water supply, the element of waste makes up part of the aggregate consumption. The causes that produce this waste are of a like nature everywhere. In the absence of preventive measures they will be fairly constant in their operation, and they will everywhere produce like effects. In the absence of definite proof to the contrary, it may fairly be assumed that wastage in New York is normal-that is, that it proceeds from the causes operative in all large cities, and is substantially equal in degree to that which arises elsewhere from similar conditions. Leaky plumbing exists everywhere; there is no reason to believe that it is more general in New York than in other cities. Defective mains and service pipes abound in all cities; and it is a violent assumption that they are more general in New York than elsewhere. The assumption that waste-prevention will accomplish more in New York than it has accomplished elsewhere may therefore be dismissed. The measure of the benefit which New York may derive from controlling the wastage of water may be approximated by a study of the results which have been reached in other cities. The most direct and practicable means of reducing waste 4 Report of Committee on Water Supply. has been proved to be the placing of meters upon all service pipes, so that all consumers shall pay in proportion to the quantity of water used. This proposition involves the outlay of several million dollars, either by the public treasury, or by private property owners. If the burden is to be placed upon the latter, general and probably successful opposition may be expected. If it is to be borne by the public treasury, it is desirable to compute the cost and the probable saving to be effected, that the net financial result may be shown. In order that this whole subject might be fully and clearly presented for public consideration, your Committee retained James H. Fuertes, Esq., C. E., to examine and report upon the probable extent of water waste in this City, practicable methods for controlling and reducing it, waste-reduction in other cities, the cost of general metering, and the financial saving resulting from utilizing the water now wasted. Mr. Fuertes's report is printed herewith in full. His essential conclusions may be summarized as follows: The readily controllable waste in Manhattan and The Bronx is about 15 per cent. of the supply, and is due mainly to leaky fixtures. It is essential to the City's safety, pending the completion of the Catskill project, that the waste be stopped, and that the water now wasted be utilized. Complete control of the present losses can only be had by universal metering. Metering of all domestic consumers will postpone for from 12 to 15 years the necessity of building the second 250,000,000 gallon installment of the new additional supply, with a net saving over the cost of metering, in interest and other charges, of about $2,000,000 a year during that period-a total saving of from $24,000,000 to $30,000,000. Considerable prejudice against general and compulsory metering of residences exists in the minds of a large part of the Prefatory public. This is due to a misapprehension of the results which follow the domestic use of meters. It is believed by many that their use will unduly restrict the reasonable and desirable consumption of water on the one hand, and impose an undue burden of expense upon property owners, on the other. Mr. Fuertes's report shows these beliefs to be groundless. Nevertheless, a very large number of property owners in this City, especially in the tenement districts, will probably oppose the introduction of meters through prejudice or misconception, unless, by the actual results of metering experimentally on a large scale, it can be shown that the prejudice against metering is not justified. In case meters are applied to all residences, the existing scale of charges should be so readjusted as to permit to every person an amply sufficient quantity of water at an annual charge that shall not exceed the present frontage rate, so that householders may be assured that the cost of water service will not, by the adoption of meter charges, be arbitrarily or unduly increased above the present annual cost. No data now available show with any degree of accuracy the normal residence consumption of water in this City, and until such normal consumption be established experimentally, it will be impracticable to determine what is a reasonable allowance per diem for each person, and upon what basis the meter charges should be adjusted. It is, moreover, desirable that experiments on a large scale be made for the purpose of learning the extent to which metering will result in stoppage of waste due to defective plumbing. The subject cannot intelligently be treated as a whole until these data have been derived from experience. In order that the data might properly be gathered, your Committee caused to be introduced into the legislature, through Hon. Mervin C. Stan' ley, the following Bill, having first submitted it for consideration to Commissioner Ellison, who expressed his general approval of the purposes thereof. The Bill having been passed by the Legislature, approved by the Mayor and signed by the Governor, has become law, and all that is necessary to give effect to its purposes is a sufficient appropriation by the Board of Estimate and Apportionment: 6 Report of Committee on Water Supply. "AN ACT "TO PROVIDE FOR OBTAINING INFORMATION AS TO THE "CONSUMPTION AND WASTE OF WATER IN THE "CITY OF NEW YORK. "The People of the State of New York, represented in Senate and Assembly, do enact as follows: "Section 1. The commissioner of water supply, gas and electricity of the city of New York, when authorized by the board of estimate and apportionment, shall proceed immediately and with all reasonable speed to designate certain districts or areas in the city of New York, which in his judgment will furnish the best opportunities for acquiring complete and reliable information as to the consumption and waste of water. "Section 2. To every service pipe not already so equipped, which furnishes water in a district designated as provided in the foregoing section, the said commissioner, when authorized thereto by the board of estimate and apportionment, shall attach without expense to the consumer, a water meter. And the said commissioner, his agents, inspectors and such other persons as he may direct, are hereby authorized to enter upon any premises for the purpose of installing meters, making examinations, or doing any other acts required to accomplish the purpose of this enactment. Section 3. The commissioner of water supply, gas and electricity shall cause inspections and readings to be made of the meters thus installed, at least once in each month and as much oftener as in his discretion may seem desirable; and he shall keep, in the office of the department of water supply, gas and electricity, a complete and accurate record of such readings, which shall be accessible to the public. "Section 4. The said commissioner shall cause to be made analyses and digests of the record aforesaid and shall report thereon to the board of estimate and apportionment, on or before the first day of January, nineteen hundred and seven, with his recommendations as to the advisability of metering all water service pipes in the city and as to the rates which should be charged to all classes of consumers of water. "Section 5. The meters thus installed, as in this act provided, shall be and remain the property of the city of New York, and shall at all times be under the supervision and control of the commissioner of water supply. gas and electricity. "Section 6. The expense of carrying into effect the provisions and intent of this act shall be borne and paid by the city of New York and to provide the means therefor, the comptroller of said city shall, when authorized by the board of estimate and apportionment and without the concurrence or approval of any other board or public body, prepare and issue corporate stock and out of the amount of money thus raised the said comptroller shall, on the certificate of the commissioner of water supply, gas and electricity, make all payments required to accomplish the purpose of this act. "Section 7. This act shall take effect immediately." 7 Prefatory A digest of existing laws relating to metering has been prepared for your Committee by Alfred L. Marilley, Esq., and is appended hereto as a separate report. An examination of these laws will show that prior to the passage of the Stanley Act, quoted above, the Commissioner of Water Supply was without power to make experimental application of meters upon a large scale, nor had the Board of Aldermen power to authorize such experimental application. It was therefore necessary to provide for the placing of experimental meters at the public expense, and to empower the Commissioner of Water Supply to enter private premises for the purpose of placing such experimental meters, no such power previously existing. The Stanley Act confers these necessary powers, and thereby enables the Commissioner to make actual tests of water consumption by all classes upon a sufficiently large scale to supply the data necessary for determining the expediency of general metering, and for the contemplated equitable readjustment of rates. Mr. Fuertes's examination has taken no cognizance of the probable increase in water revenues which will result from general metering. It has been shown by experience that many individual premises in the classes now subject under the law to meter measurements, have eluded the payment of meter charges, and have consumed excessive quantities of water without making due payment therefor. It is reasonably certain that general metering will largely increase the revenues from business and manufacturing concerns who now use water without proper payment. This probable result should be considered as one of the reasons for general metering, and it would probably add largely to the financial advantage outlined by Mr. Fuertes, who has considered only the financial saving to the City to be effected by making more distant the time when a large investment would be required for additional water supply. In addition to the technical and legal reports by Mr. Fuertes and Mr. Marilley, it has seemed to your Committee desirable to append to this report a concise summary of the general conditions relating to the water supply of this city. Such a summary has been prepared, at the request of the Committee, by J. Report of Committee on Water Supply. Hampden Dougherty, Esq., former Commissioner of Water Supply, Gas and Electricity, and is appended hereto. It is of the first importance that the people of this City apprehend clearly the great danger to which the population is subjected by reason of a possible, and even probable, water famine pending the completion of the new supply; and that such danger is only to be averted by repressive measures which, by reducing waste and leakage, shall utilize the present supply to its fullest capacity. In the opinion of your Committee general metering is the only practicable safeguard, and it is this Committee's purpose, therefore, to urge upon the public authorities that recourse be had to that measure as soon as possible. The Stanley Act will afford means for gathering all necessary data, and for determining the proper readjustment of the charges. It is hoped and believed that the information gathered under the Stanley Act will be so convincing to the people, and to the public authorities, as to enable the passage of an act by the next Legislature that shall permit the prompt adoption of general metering upon a basis that will meet with general public approval. It is the intent of your Committee to pursue this subject diligently, and to keep closely in touch with the work that may be done by the Water Department under the Stanley Act. All of which is respectfully submitted. COMMITTEE ON WATER SUPPLY, THE MERCHANTS' ASSOCIATION OF NEW YORK, By Henry R. Towne, Chairman. NEW YORK, June 18, 1906. Ir CI WASTE OF WATER IN NEW YORK AND ITS REDUCTION BY METERS AND INSPECTION BY JAMES H. FUERTES, C. E. PAGE WASTE OF WATER IN NEW YORK AND ITS REDUCTION BY METERS AND INSPECTION..... 13 APPENDIX A: UNACCOUNTED-FOR WATER... 107 APPENDIX B: WASTE OF METERED WATER... 113 APPENDIX C: LEAKAGE FROM STREET MAINS.. 117 APPENDIX D: WASTE-REDUCTION IN AMERICAN AND GERMAN CITIES....... 127 APPENDIX E: LIST OF REFERENCES.....225 WASTE OF WATER IN NEW YORK AND ITS REDUCTION BY METERS AND INSPECTION. NEW YORK, May 9, 1906. Mr. Henry R. Towne, Chairman Committee on Water Supply, The Merchants' Association of New York. SIR: IN accordance with your request I have the honor to present a statement of the results of an investigation conducted under the general direction of your Committee for the purpose of ascertaining, as nearly as possible: 1.-The amount of water lost by leakage and wastage from New York's Water Supply. 2.-The proportion of water so lost that could be quickly and effectively saved. 3.-The bearing which the saving of this lost water would have on the advisability of deferring the construction of the works for the proposed new Water Supply from the Catskill Mountainis. SUMMARY OF CONCLUSIONS. BRIEFLY stated, the results of this investigation, which has been carried as far as the time available would permit, may be summed up as follows: In Manhattan and the Bronx the amount of leakage and wastage effectively and quickly controllable probably did not average for the ten years prior to 1904 more than about 15 per cent of the supply, and was largely confined to the fixtures and services of unmetered consumers. (See Parts I, II, III, IV, V, and Appendices A, B, and C.) 13 Waste of Water and Its Reduction. The leakage from the street mains is relatively small in amount and such as exists could not be stopped short of many years of costly work. Its complete stoppage, even if possible, would be of very little benefit as compared with the control of the leakage and wastage taking place from the fixtures on the premises of unmetered consumers. (See Appendix C.) Manhattan and the Bronx are at the present time not only using more water daily than can be safely depended upon from the Croton and Bronx water-sheds, with the present available storage, but are actually using within about 17 per cent of the greatest amount that could be continuously obtained if the reservoirs were large enough to hold back for use the entire stream flow, allowing no water to waste over the dams during pretty nearly a generation. (With meteorological conditions identical with those which have prevailed since 1879.) (See Part VI.) Only 17 per cent more water could be continuously had from the Croton water-shed, no matter how much storage were provided, for the reason that the City is now using within 17 per cent of the average flow of the Croton River. To get this last 17 per cent the storage would have to be about two and one-third times as great as the combined capacities of all the existing reservoirs, lakes and ponds now on the water-shed. Such storage can not be had; therefore New York must go to other water-sheds for more water. (See Part VII.) If the storage in the Croton water-shed is increased materially beyond the present amount it will be at the risk of spoiling the water. (See Part VII.) During recent years the relative amount of rain falling in the different seasons has been favorable to large summer stream flows in the Croton water-shed. If in any year since 1901 the same total amount of rain had been distributed differently, that is: if more had fallen in the Summer and less in the Spring, under conditions no less extreme than have frequently prevailed in the past, the Water Department would have been compelled to partially shut off the supply for several months at a time to avoid emptying the storage reser14 Summary of Conclusions. voirs. Cutting off a part of the supply would reduce the pressures all over the City with the added risks of general conflagrations (due to the inability, through lack of water, to control small fires). (See Parts VI and VII.) The flow of the Croton River has been above the average for several years. It must soon, therefore, fall below. It requires no exaggeration to portray the dangers of the situation when it is realized that the amount of water the City is using from the Croton water-shed is within 17 per cent of the average flow of the Croton River and that the flow of this river, as known by its past records, may possibly fall for an entire year to an amount 40 per cent below the average. (See Part VII.) With the greatest amount of waste and leak reduction possible of practical accomplishment during the coming years, and with the construction of the proposed Cross River reservoir and all other emergency reservoirs that can be built in the Croton water-shed, the City will not be out of danger of a waterfamine until the new supply is made available. The construction of the first installment of the new works should therefore be started immediately and should be pushed to completion as rapidly as possible. (See Part VII.) In order to carry the City through the intervening years that must elapse before the new supply can be built, with a reasonable assurance of averting the probable certainty of a short supply (if the lessons to be learned from the record of the flow of the Croton River for the past 38 years can be depended on as an indication of future conditions), comprehensive steps to stop leakage and wastage must be taken immediately and must be maintained without relaxation. (See Parts I, VII, VIII, and IX.) Leakage and wastage can be controlled by house to house inspections, or by inspections in conjunction with partial metering, or by universal metering. The control of leakage and wastage by house to house inspections will be incomplete, as, in order to have been detected, leaks 15 Waste of Water and Its Reduction. must have already existed. Relatively complete control of these losses can only be had by universal metering. The metering of all domestic consumers will prove a very profitable investment in New York City by postponing for from 12 to 15 years the necessity of building the second 250,000,000 gallon installment of the new additional supply, with a net saving over the cost of metering of about $2,000,000 per year for that period. (See Part IX.) As it is imperative that as great a reduction as possible be immediately secured in leakage and wastage, the conditions demand: First.-The organization and placing in effect at once of a comprehensive system of house-to-house inspections, coupled with examinations of the flows in the street mains (for the purpose of locating quickly the most serious losses) to carry the City through until the question of final control of wastage can be settled. Second.-There must be secured to the City the right, which it does not now possess, to place meters on the services of domestic consumers. (See Parts I, VII, VIII, and IX.) A brief digest of the data upon which the foregoing statements are based will be found in the following pages. 16 PART I. METHODS FOR CONTROLLING LOSS OF WATER. IT is my opinion that the term wastage as applied to the various losses of water in a city's supply has been extremely unfortunate, implying more than it really should. The proportion of water actually wasted, in the ordinarily accepted meaning of the word, is relatively small. Most of the lost water is leakage, not wastage; and a campaign, to prevent this loss of water, which is a loss of wealth, should be directed against leakage. I believe that the number of persons who deliberately waste any significant amount of water in New York is probably no greater than the number who steal it through illicit connections. On the other hand the number who allow it to leak continually from defective fixtures on their premises is probably equal to the number of unmetered users. This is not maliciousness, but heedlessness, carelessness, shiftlessness, for which, not the people, but the system under which the water is distributed and sold is responsible. Owing to the erroneous but very general impression that waste reduction implies a restriction in the use of water, it has become extremely difficult in large cities to inaugurate such reforms as would quickly lead to material financial benefits not only to the city as a whole but to the individual owners of properties, without in any way depriving a single citizen of the tse of as much water as he can use, or has been accustomed to using. In a city the population is broadly divisible into two general classes: owners and tenants, the tenants being in number largely in excess of the owners. In New York the water taxes fall directly upon the owners, the tenants, except in relatively small numbers, having no direct interest in the question of the suppression of leakage, as the water rates are supposedly covered by the rent. The owners of properties rented to tenants are therefore the ones who control the situation for the reason that by the exertion of political pressure, and by threats of increased rents, it is 17 Waste of Water and Its Reduction. entirely possible for this relatively small body of citizens to defeat action looking towards a material betterment of conditions. The problem, therefore, becomes one of convincing the owners that it is to their interest to assist in stopping the leakage which is adding materially to their taxes year by year. To accomplish this purpose it is necessary to show: First.-That leakage can be stopped quickly; and that when once stopped it can be permanently controlled. Second.-That owners of rented properties run no risk of being robbed by wasteful tenants when the water is sold by measure instead of being supplied at frontage or assessment rates. Third.-That it is to the financial advantage of the owner to keep the plumbing in his buildings properly repaired and pay for the water used rather than to neglect to keep his property up and have it saddled with increased taxation to cover the value of the significant percentage of the city's water now running continuously into the sewers. If there were only a few shiftless landlords in the city their carelessness would not greatly affect the total supply, but unfortunately each owner labors under the delusion that the value of the water lost on his premises is so small that when charged to all the other property in the city, as it must be, his individual contribution to cover the wastage will be insignificant in comparison with the cost of stopping the leaks. With every other owner pursuing the same policy, however, each individual is helping to pay for the shiftlessness not only of himself but of every one else, with the net result that the water taxes must be made large enough to cover the value of the water thus allowed to leak away. House-to-house inspections coupled with the testing of mains to locate leakage and wastage, will, if repeated at least twice a year, control a variable percentage, probably about three-fourths, of the total controllable loss. In order to be of any value this plan would have to be maintained in operation perpetually; its discontinuance would be immediately followed by the resumption of leakage and wastage, no permanent benefits would have been obtained and the funds appropriated would have been expended on a temporary expedient. 18 Part I: Methods of Controlling Loss. It has been the universal experience of every city in the United States that wastage and leakage can not be satisfactorily controlled without the use of water meters on practically every service, domestic, manufacturing, commercial, and public. Whether the city be put entirely on a meter basis, or whether the meters be used merely to locate wastage and leakage, these losses being stopped by the infliction of penalties for second offenses, I believe to be a subject for further investigation and discussion, so far as controlling wastage is concerned. But there is no doubt but that in American cities, where administrations change frequently and where appropriations must be asked of councilmanic committees the meter must be used as an adjunct in waste prevention if an adequate return is desired for the investment. Further, the metering of a relatively small proportion of the domestic consumers may be of no avail unless these meters be placed on wasteful consumers who have been located by inspections, in which case probably the greater part of the waste and leakage can be controlled with the metering of about 15 per cent of the consumers; but even thereafter the inspections will have to be kept up persistently and be followed by the installation of other meters, until finally, before the situation is well in hand, practically all the services will have been metered. When that time arrives, and not until then, will the wastage and leakage question cease to be one of great moment in this. city. The records of every city in the world where meters have been used in sufficient extent show that the amount of unaccounted for water can be brought down to and maintained at a reasonable limit; and these records further afford convincing proof of the proposition that after meters have been placed on every service wastage will be kept permanently under control. In every municipal Water Supply System the question of combatting wastage and leakage comes up sooner or later for settlement. Sonie cities have solved the problem successfully; others have spent large amounts, thereby securing temporary benefits only to find some years later that conditions were as bad as ever. In every city where satisfactory results have been obtained and permanently secured the campaigns have been conducted against the wastage in the premises of all consumers, domestic, manufac19 Waste of Water and Its Reduction. turing and public. In New York the manufacturing consumption being now metered can not be reduced to much lower figures than at present obtain. The discovery of illicit water takings, and testing of meters, however, will tend to increase the revenues of the Water Department and may lead to the discontinuance of some of the illicit water takings or the suppression of wastage therefrom. The amount of wastage from such connections can not be known nor even guessed at from any data now available, but it seems certain that its amount must be insignificant in comparison with the wastage from the hundreds of thousands of leaky faucets, ball cocks and other fixtures to be found in the buildings, public and private, throughout the entire city. The most reliable guide by which to judge of the effectiveness and permanency of methods for reducing wastage is the actual experience of other cities in which this problem has been attacked; and when the various methods tried have been studied in the light of local conditions the reasons for failures or successes become clearly apparent. Fortunately, a great deal of such work has been done both in American and foreign cities and there are ample facts to be had from which a reliable forecast may be made of what may be expected as the result of following certain policies. Generally speaking there are but two methods of stopping wastage: one is to maintain a bureau for systematic house-tohouse inspections, frequently repeating the inspections each year, and inflicting penalties on those who permit leakage from fixtures on their premises, coupling with this systematic examinations of the flow in the street mains to locate leakage from underground service pipes, which leakage would not otherwise be detected. The second method is to sell the water only by measurement instead of supplying it at frontage or assessment rates. The first system may be productive of excellent results for so long a time as the method is vigorously followed up. As soon as it is relaxed matters quickly become as bad as before and unless temporary results only are desired the expense incurred will have been, in effect, a wastage of the public funds. The second method wherever it has been employed has proved 20() Part I: Methods of Controlling Loss. a satisfactory means of holding leakage down to the lowest practicable amounts. The various cities which have undertaken the control of leakage and wastage may for a convenient discussion of the results secured be divided into four general classes: A.-Cities selling water at assessment or frontage rates and in which waste reductions have been accomplished by house-tohouse inspections, metering wasteful consumers, or combinations of these methods, and in which, when these methods have been relaxed, the good results obtained have been subsequently lost. B.-Cities in which a considerable proportion of the services are metered, effecting some improvement, but in which wastage and leakage are not yet entirely under control. C.-Cities in which waste reduction has been successful and in which the good results secured have been permanently maintained. D.-Cities in which water has been sold by measure for a good many years and in which wastage and leakage have been and are now under as perfect control as practicable of attainment. For the purpose of illustrating the effect of different methods of handling this question it has seemed advisable to describe briefly the actual experiences of a few cities representative of each of the above classes. These descriptions have been prepared from data compiled from the Annual Reports of the Water Boards of the various cities, and from correspondence with water works officials. It has been stated that there are really but two practicable methods of controlling wastage. This I believe to be true. Temporary reduction of the pressure in the mains is sometimes resorted to as expedient to tide over brief periods of shortage. New York has already been through several experiences of this sort. In connection with waste reduction proper plumbing regulations are a necessity, particularly when inspections are depended upon to detect wastage. In Manchester and several of the larger English cities it has been for many years the custom for the city to supervise the plumbing very closely, testing every fixture before it is allowed to be used and stamping it with an official seal 21 Waste of Water and Its Reduction. if satisfactory in make, material and style, and properly passing the prescribed tests. In one small English town where the plumbing had become so hopelessly bad that repairs seemed impossible the water department practically installed new and satisfactory fixtures in every house before attempting to control leakage and wastage. In Great Britain water meters are not extensively employed, nearly all the water used for domestic purposes being paid for on the basis of the rental value of the property occupied by the consumer instead of being based, as is commonly done in America. on frontage rates with additional charges for the extra fixtures in use; and the control of wastage and leakage is effected by frequent, thorough, house-to-house inspections coupled with the use of only approved plumbing fixtures. In these cities, however, the municipal organizations are very different from those in our American cities. With our political complications arising from frequent changes of administration it has always been impossible to accomplish lasting reforms, in the matter of waste suppression, by means which have been satisfactory in other countries. With constant changing of the heads of departments and of the members of council, and committees on whom reliance must be placed for appropriations, reforms inaugurated by one administration may be discontinued by the one succeeding and probably before any substantial benefits may have been secured. This is the condensed history of waste reduction by inspection in the American cities which have tried it. New York has been through it twice, the successful campaigns having been conducted under the direction of Allen Campbell from 1876 to 1880 and his successor, Hubert O. Thompson, from 1880 to 1884, and General John Newton from 1886 to 1889. In the intervening period and since the time of General Newton, no aggressive, systematic work of this nature has been carried on and the use, wastage and leakage have steadily increased during the last 15 years until the total consumption of water per person, in New York, is now nearly double what it was when General Newton's term of office ended. A.-Cities in which waste reduction has been accomplished by house-to-house inspection, metering wasteful consumers, or both, and in which these methods have been relaxed, showing reduction of wastage and subsequent return of previous wastage. 22 Part I: Methods of Controlling Loss. AMONG cities where extensive work in waste suppression has been done, Boston, Detroit, Newark and St. Louis, all afford useful data. Table 1 contains a condensed summary of the statistics of these cities. It will be observed that in each of these the average daily consumption per capita has been materially reduced only to increase again in subsequent years. The reasons for this will be found in the detailed descriptions of the methods used and the results obtained. TABLE NO. 1. Year Average Percentage CITY. Referred Daily Con- of Services Years to. sumption Metered. Elapsed. Per Capita. Newark................... 1898 113.0 10.0.................. 1901 94.0 32.0 3................... 1904 105.0 42.0 3 Detroit.................... 1889 210.0 0.6.................... 1897 136.0 18.0 8 "................... 1905 181.0 27.0 8 Boston..................... 1883 91.5......... 1884 68.0..... 1 1892 95.0..... 8 1896 101.0..... 4.................... 1900 113.0 11.8 4 The most extensive work ever done in the United States in waste reduction by means of house-to-house inspections, combined with examinations of the flow in the street mains to locate wastage, was undertaken by the City of Boston in 1880 and prosecuted vigorously for several years. This was the first and only extensive application in the United States of the system devised by Mir. George F. Deacon, City Engineer of Liverpool, and subsequently used as a permanent institution in that city, in Glasgow, Manchester, and many small English towns. So far as the writer has seen any references thereto it has never been permanently adopted outside of England, although used for limited times in Frankfort-on-Main, in Yokohama, and for a short time in the American cities-Detroit, Mich., 1896; Philadelphia, 1898; Washington, D. C., 1899. Waste of Water and Its Reduction. As a result of the very thorough work done in Boston under the direction of Mr. Dexter Brackett the consumption was cut down from 91.5 gallons per capita in 1883 to 68 gallons in 1884. During the succeeding years the thoroughness with which the work was prosecuted was gradually relaxed until in 1895 it was temporarily abandoned owing to its excessively heavy cost and the opening of the new Metropolitan Water Works which supplied an abundance of water and made the control of wastage less imperative. Since 1895 the system has again been used more or less intermittently but not with the thoroughness of the first application under Mr. Brackett's direction. The cost was very considerable, and although recent detailed figures are not available, it is safe to say that after a total expenditure of more than one million dollars on waste reduction conditions in this regard are no better to-day than before this money was expended. This is not to be in any way interpreted as an implication that the money was not wisely spent; but merely as an illustration that this system of waste prevention, which is burdensome and disagreeable to householders, as well as expensive and difficult to maintain in a state of efficiency, will produce no permanent, lasting results unless pushed vigorously and continuously by a large corps of skilled inspectors under the direction of an energetic, capable director backed by a strong support from the administration, and supplied with ample funds. The same system applied to New York and Brooklyn would cost not less than $1,000,000 per year, if it were possible to use it; and it would cost not less than $1,000,000 to install, as satisfactory results could not be obtained without putting curb-cocks on every service pipe, installing numberless valves in the mains and tracing out obscure connections of which there are now no records. B.-Cities in which a relatively large number of meters are in use, but in which wastage is not yet under control. A S a general thing wastage and leakage are confined at any given time to a relatively small proportion of the consumption. If it were possible by any practicable means to always find 24 Part I: Methods of Controlling Loss. these few careless or extravagant users wastage could be controlled by inspections and penalties. Unfortunately some premises where no leakage or wastage takes place to-day may be among the worst offenders to-morrow. Again, in some cities a large amount of water is wasted through the fixtures in public buildings, through fountains and watering troughs, or needlessly wasted by extravagant use for street sprinkling, sewer flushing and other purposes for which a moderate quantity would be ample. The general proposition that metering wasteful consumers will control wastage, or that metering a certain percentage of the services will be an effective check on wastage, can only be true when the meters are shifted around from place to place as the necessity therefor is made apparent by inspections. Furthermore, this expedient can never be more than partially effective for the reason that wastage must already have taken place before the necessity for stopping it will have become apparent. TABLE NO. 2. CITIES HAVING A CONSIDERABLE PROPORTION OF THE SERVICES METERED, BUT IN WHICH WASTAGE IS NOT YET UNDER CONTROL. Average Year Daily Con- Percent- Inhab- Number CITY Referred sumption age of itants per of to. per Capita, Seervces Service. Years Gallons. Metered. Elapsed. Cleveland............. 1891 136**....... 1901 174** 6 7.5 10............. 1904 138** 49 7.5 3 Richmond............. 1899 126 30 5.8............ 1904 129 41 6.1 5 Springfield............. 1893 117** 17 5.3............. 1899 113** 20 5.0 6............. 1903 216** 36 5.1 4............. 1904 130** 37 7.0 1 Syracuse*.............. 1898 81 44 7.9.............. 1904 102 72 7.3 6 Taunton.............. 1898 48** 38 6.7.1............. 904 64** 45 5.6 6 * In Syracuse there is an extraordinarily large number of watering troughs which alone waste about twice as much water as should be required for all public purposes. There is a large amount of waste from fountains and public buildings also. ** Per consumer. 25 Waste of Water and Its Reduction. In Cleveland; Richmond, Va.; Springfield, Mass.; Syracuse, N. Y., and Taunton, Mass., a considerable proportion of the water for both manufacturing and domestic purposes has been sold by meter measurement for several years, and as it can not be said that in any of these cities wastage is under satisfactory control a brief statement of the conditions, methods used and results obtained will prove profitable., Table 2 contains a condensed statement of the results in each of these cities, with the number of inhabitants per service, for the years given, as an indication that the proportion of the population actually supplied during these years was sufficiently constant in each city not to materially affect the per capita rate of consumption on which the comparisons are based. C.-Cities in which waste reduction methods have been successful and in which the results obtained have been permanent. IN all the fore-mentioned cities, in which waste reduction has been practised by house-to-house inspections, partial metering, combined inspection and metering, and one or both of these expedients combined with testing of the flow in the street mains for the location of leaks and of wasteful consumers, it will be noted that although in several cases the reductions in wastage have been considerable in not one has the benefit obtained been permanently secured. In the cities listed in Table 3 not only has success been attained in the suppression of undue leakage and excessive wastage, but these results have been continued from year to year with no tendency to a return of former conditions. 26 Part I: Methods of Controlling Loss. TABLE NO. 3. CITIES IN WHICH WASTE REDUCTION METHODS HAVE BEEN SUCCESSFUL, AND IN WHICH PERMANENT RESULTS HAVE BEEN SECURED. Reduction of P Time Wastage. Per- me Average Wastage. cent- in Year Daily Con- Gals. age which CITY. Re- sumption per of Ser- Reduc- Results. ferred per Capita. Capita Per- vices tion to. Gals. per cent- Me- was seday. age tered. cured. Years. Asb'y Park,N.J.tt 1900........... 50. 1901......... 50 100. 1 Permanent Atlanta, Ga...... 1884 225t*..... 30....1888 91.0 134 60 100. 4 Permanent Atl'tic City,N.J.tt 1896........ 52. 1898....... 60 71. 2 Permanent Fall River, Mass.. 1874 84.5.. 8...1884 32.5 52 62 67. 10 Permanent Harrisburg, Pa... 1891 89.0*... 1897 60.0 29 32.. 6 Permanent Hartford, Conn.. 1899 110.0.. 6... 1904 70.0 40 36 94. 5 Permanent Lawrence, Mass... 1892 87.0... 39...1904 42.0 45 52 87. 12 Permanent Lowell, Mass..... 1900 83.0.. 52......1903 51.0 32 39 65. 3 Probably permanent Madison, Wis.... 1885 129.0**.... 0.5......1898 50.0 79 60 92. 13 Permanent Milwaukee, Wis.. 1894 108.0.... 47... 1898 85.0 23 20 61. 4 Permanent Poughk'psie, N.Y. 1876 1,472.0**.... 4. 1884 857.0.... 18. "1886 1,023.0... 25. 1889 799.0.... 41. " 1892 814.0.. 58. 1894 614.0 858 58 66. 18 Permanent Wellesley, Mass... 1890 96..O. 18.. 1895 58. t 38 40 100. 5 Permanent tt Summer resorts; small residential, but large summer population; no per capita figures possible. Reduction based on yearly pumpages. * Domestic, public and unaccounted for water. (Manufacturing consumption deducted from total.) ** Based on 6.06 persons per service. t Consumption per consumer. t * Based on 7.65 persons per service. *t* Consumption per service. Waste of Water and Its Reduction. D.-Cities in which large percentages of services have been metered for many years, showing permanent control of wastage. N addition to the cities already mentioned in which waste reduction has been successfully accomplished, there are several in which wastage has been under complete control for numbers of years. A list of a few of these is given in Table 4. TABLE NO. 4. CITIES IN WHICH LARGE PERCENTAGES OF SERVICES HAVE BEEN METERED FOR MANY YEARS, SHOWING PERMANENT CONTROL OF WASTAGE..... CITY. Atlanta................ Fall River............... Harisburg............ Madisonurg.............. Milwaukee............... Madison.................. Milwaukee................ Poughkeepsie............ u Providence............. Wellesley................ Yonkers*................. M............... Average Year Daily Referred Consumption to. per Capita. Gallons. Number of Years Elapsed. Percentages of Services Metered. 1888 1904 1884 1902 1897 1904 1896 1904 1898 1904 1889 1892 1894 1899 1905 1891 1894 1900 1904 1895 1904 1900 1902 1904 -I - I 91 76 33 40 60 65 67 71 85 89 799t 814 614 635 586 51 63 54 67 58 55 78 89 94 I 16 18 7 8 6 3 2 5 6 13 9 4 All All 67 95 60 65 79 96 61 80 41 58 66 80 86 63 69 83 86 100 100 98 96 100 I * Yonkers' supply is all pumped and no allowance is made in figures given for slip of pumps. The Yonkers records run back many years prior to the first date given. t Per service. Before leaving this subject it seems desirable to point out, in view of the comparatively few long-time records available for 28 Part I: Methods of Controlling Loss. American cities, that what is true in regard to leakage and wastage, methods of controlling these losses and results obtained by different methods of control in the United States, has been found to be true in other countries. The best foreign data on this subject can be had from Germany; and the summaries of the statistics of the several German cities quoted in Tables 5 to 11, inclusive, will serve to show that the problems in water works management presented in Germany years ago are those that are fast becoming of great economic importance in the United States. HE conditions in German cities are so very different from those in America that no direct comparisons between the relative amount of water consumed can with propriety be made. Nevertheless waste is prevalent in Germany, as it is wherever there are people, and the experience of the German water works managers in controlling this wastage are instructive, illuminating, and worthy of careful attention. The great detail in which records have been kept to account for the water supplied to the German cities makes the data particularly valuable. It is almost impossible to secure from our American cities data in sufficient detail to be of much value. Rarely is it possible to find statements of the percentage of the supply sold by measure, the proportions used for public purposes, for the purposes of the plant, for flushing out the street mains, for extinguishing fires, for water used in public buildings, comfort stations, public fountains, park uses, street washing and street sprinkling; and almost no records can be found in which the difference between the reported consumption and the accounted-for water can be considered an accurate measure of the loss by leakage from the mains and pipes in the streets. Of course it must be understood that conditions in German cities are entirely different from those at home, not only in regard to the requirements of the people, their mode of living, etc., but as to the laws under which municipalities are governed. In practically all the older German cities by far the greater proportion of the people live in apartments or flats, detached houses being few in number. For this reason the population is usually very great per unit of area, which circumstance contributes largely to the small per capita consumption of water. Many of 29 Waste of Water and Its Reduction. these cities were in former times surrounded by walls for protection. In some cases these walls are still standing, in whole or in part, and increased population has been taken care of by overcrowding. Until comparatively recent years most of the German cities, many of which were very old before America was discovered, possessed no cheap means of passenger transportation, and workingmen and women were compelled to live closely enough to their places of business to be able to go back and forth on foot without too great loss of time; and as trade and commerce increased in volume the value of property became greater, necessitating larger rentals, which, in turn, could only be satisfied by crowding more and more people into each building. This is the origin of the flat system in these cities, and even now, when in many the facilities are equal to if not surpassing our own in certain respects, the old customs still prevail, being more natural, and undoubtedly more economical. It is largely due to this custom that German cities have so low a per capita consumption of water. The number of fixtures from which water can leak and waste is reduced to a minimum; the lengths of distribution pipes necessary to supply the water, and consequently, the numbers of opportunities for losses from leaky joints, hydrants, valves, service pipes, etc., are reduced to a minimum; the difficulties of keeping close watch to prevent leakage and wastage are reduced to a minimum and consequently the expenses of administration. Contrast the conditions in Berlin with those in Detroit for instance. In Berlin there is an average of over 70 people supplied by each service pipe; in Detroit a little over five people. The opportunities for losses and leakage at Detroit, therefore, are at least 14 times (in reality much more than this) as great as in Berlin and a loss of 10 per cent of the water at Berlin by leakage, representing 1.8 gallons per capita per day, would be no less extravagant than a loss of 25 gallons per capita per day in Detroit, assuming that in Berlin there were as great a number of fixtures to each family as in Detroit. In 1890, exclusive of those of a public nature, there were in Berlin about 17,000 bath tubs for a population of 1,396,963 people, or one tub for every 79 people. When also it is understood that, Part I: Methods of Controlling Loss. excepting in the finer hotels and better class of apartments, permanent wash stands with hot and cold water, and wash tubs for laundry purposes supplied with hot and cold water, and such other fixtures as are common in our cities, are unknown in most of the European cities, it will be realized that the opportunities for wasting water are limited. And it will also be apparent that comparisons between American cities and European cities which do not take these conditions into account are misleading and improper. The following statistics are given, therefore, not for the purpose of comparing the German with the American water consumption, but for the reason that they point out that even where the demands for water are very small, as measured by our home standards, waste takes place and is sensitive to remedial measures in almost as great a proportion as in our American cities. In other words, although the wastes are smaller because the opportunities for waste are smaller, in proportion they are as great as those to which we are accustomed; and so far as the personal factor in waste is concerned, the American, as an individual, is not more wasteful than his cousins across the seas. More water has to be supplied to the American, the conditions existing in the cities in which he lives demanding it, and he consumes and uses more water; but the waste, or, to express it more properly, the lost water, is not chargeable to the wastefulness of the individual, but to the fact that in our cities the opportunities for leakage from mains and service pipes and from faulty plumbing, old, worn-out faucets, ball-cocks, etc., are multiplied many fold. The data from which the statistics for the following cities were compiled were derived mainly from the Statistical Summary of the German Water Works, published by the German Society of Gas and Water Experts, arranged and edited by E. Grahn, under the title, "Die Stadtische Wasserversorgung in Deutschen Reiche sowie in einigen Nachbarlandern," printed by R. Oldenbourg, Munich and Berlin, 1902. This monumental work contains in great detail data regarding nearly all the water works plants in the German Empire and neighboring countries, and is a model, in many respects, of the form in which statistical data regarding water works plants should be recorded. 31 Waste of Water and Its Reduction. The cities chosen for illustration represent a combined population of over 7,000,000 people and were selected almost at random. Many well-known cities were omitted for the only reason that reproducing their statistics would simply be repeating the same story over and over. In the following tables the statistics are summarized in groups according to the percentage of the consumption passed through meters. The first group, Table 5, contains 25 cities in which practically all water is sold by meter. It will be observed that each of these completely metered cities has its own characteristic rate of consumption; but the variation from lowest to highest exhibits as great a percentage range as will be found in any country. 32 Part I: Methods of Controlling Loss. TABLE NO. 5. CITIES IN WHICH ALL CONSUMERS ARE METERED. Per Capita Consumption. Percentage Year Gallons per Day. of Supply CITY. Re- Population. ferred Conto. Conto. Total. sumers Public Plant Conalone. Uses. Total. sumGriez.......... 1899 23,000 7.6 6.9 0.7... 90 100 Quedlinburg.... 1896 22,000 8.2 5.7 0.6 1.9 76 100 Gotha.......... 1899 31,670 10.3 6.3 1.6 2.4 61 100 Potsdam....... 1896 55,849 12.7 10.1 1.1 1.5.. 100 Erfurt......... 1896 78,167 13.5 10.5 3.0... 77 100 Eisenach....... 1899 33,886 14.3 9.5 3.4 1:4 67 99 Weimar........ 1897..... 14.5 9.3 1.5 3.7 74 100 Rudolstadt..... 1899 13,000 15.1 10.4 0.3 4.4.. 100 Bamberg....... 1897 39,500 15.8 9.8 3.3 2.7 47 78 Charlottenburg. 1896 350,000 16.1........ 85... Offenbach...... 1900 50,500 17.1 9.1 3.1 4.9.. 100 Mainz.......... 1900 80,000 17.2 10.3 3.6 3.3 75 100 Berlin.......... 1896 1,719,183 18.2 14.9 1.6 1.7 82 100 Colmar......... 1900 40,000 18.8 8.5 8.6 1.7.. 100 Stettin......... 1896 140,000 20.0 18.6 1.2.2 76... Hannover...... 1896 246,000 21.2 15.7 2.0 3.5 76 100 Giessen......... 1900 25,200 21.2 16.1 3.4 1.7 90 100 Heidelberg..... 1899 38,000 21.4 14.4 6.4.6 66 100 Stralsund...... 1896 30,105 21.4 13.0 1.6 6.8 60 100 Darmstadt..... 1900 70,000 21.4 18.8 0.5 2.1 88 100 Mannheim...... 1900 120,800 22.5 16.7 2.4 3.4 74 100 Breslau........ 1896 370,000 23.2 17.4 1.8 4.0 78 100 Magdeburg...... 1896 224,235 24.8 20.0 1.1 3.7 86 100 Strassburg...... 1900 117,500 27.2 13.3 11.2 2.7 49 100 Worms......... 1900 32,000 28.8 24.2 2.6 2.0 88 100 To show the lack of correspondence between the percentage of the consumption metered and the per capita consumption, Table 6 has been prepared. This contains the statistics, arranged in pairs, of cities of about the same population, and having the same per capita rate of consumption but widely varying percentages of the supply metered. It would seem hardly necessary to make such comparisons, but the argument is so frequently used by the layman that because such and such a city is metered and has a low rate of consumption, therefore such and such cities, if metered, should also have their rates of consumption reduced to the same amounts, that it seems advisable to refute such statements decisively by the actual results of experience. 33 Waste of Water and Its Reduction. TABLE NO. 6..... CITY. Frankfort O-M......... Eisenach............... Kiel................... Darmstadt............. Danzig................ Mannheim............. Bremen................ Magdeburg............. Bernberg.............. Worms............... Year referred to. 1895 1899 1896 1900 1895 1900 1900 1896 1900 1900 Population. 59,049 33,886 86,000 70,000 120,000 120,800 167,000 224,235 34,500 32,000 Per capita Consumption Gallons Daily. 14.3 14.3 21.9 21.4 23.7 22.5 25.7 24.8 27.5 28.8 PERCENTAGE OF Total Consumers Supply Supply Metered. Metered. 43 67 99 10 88 100 44 74 100 28. 86 100 8 88 100 Tables 7 and 8 contain the records of cities partially metered, and cities practically without meters. In Table 7, exhibiting the statistics of unmetered cities, will be found several with rates of consumption no higher than those either partially or completely metered. The significance of this is that meters have no influence upon the amount of water required. Careful inspections for wastage, maintained systematically without relaxation, with the enforcement of penalties for serious infractions of the regulations of the water departments are nearly as efficacious as meters in restricting wastage and have the same result: viz., to prevent waste, but not use. On the other hand, in unmetered cities where no care is taken to husband the supplies the consumption may reach very high amounts. It is not possible from the data at hand, and without intimate knowledge of the conditions in each city, to criticise or even explain the reasons why some cities have high and some low rates of consumption. 34 Part I: Methods of Controlling Loss. TABLE NO. 7. ARE PRACTICALLY UNMETERED. CITIES WHICH PER CAPITA CONSUMPTION, GALLONS PER DAY. CITY. Kiel...... '' Bremen...... Bernberg..... Karlsruhe.. Dessau...... Regensburg.... Metz....... Wiirzbur~g..... Augsburg..... Luabeck...... Year. 1896 1900 1900 1899 1900 1897 1900 1897 1895 1900 Population. 86,200 167,000 34,500 95,000 50,000 47,400 54,517 71,000 81.,900 73,093 Total. 21.9 25.7 27.5 32.0 33.0 34.9 35.7 56.6 63.8 82.8 Consurners Alone. 18.2 23.7 28.5 13.0 34.5 56'.0 Public Plant. Uses. 2.4 1.3 3.3 0.5 3.5.. 6.6 16.1 18.7 3.4 7.8.. Percentage of supply Metered. 10 28 8 30 17 18 12 6 1 9 TABLE NO. 8. CITIES WHICH ARE PARTIALLY METERED. PER CAPITA CONSUMPTION, GALLONS PER DAY. PercentCT. Ya.Popula- - ___-___ ___age of tion. Con- Pub- MeTotal. surners lic Plant. tered. Alone. Uses. Frankfort O-Ml 1895 59,049 14.3 9.7 4.4.2 4 3 Koenigsberg.. 1896 172,000 17.4 15.1 1.4.9 4 1 Leipzig.......1898 422,000 17.7......... N-urnberg......1897 177,000 20.3 13.0 2.6 4.7 63 Hle1896 117,500 21.7 20:'3 1.3 0.1 78 Danzig.......1895 120,000 23.7......... 44 Dresden......1898 366,000 26.7........ Muilihausen.....1900 84,000 27.0......... 64 Hamburg. 1899 684,400 47.0 44.0 3.0.... 49 For convenience the data regarding these cities have been rearranged in the form of summaries for each group. Table 9 contains the data regarding waste reduction in the cities which formerly sold water on the assessment plan but changed to the plan of sale by measure. Table 10 contains the data regarding cities in w'hich a comparatively small percentage of the consump115 Waste of Water and Its Reduction. tion is sold by measure, and Table 11 the data regarding cities in which practically all the water has been sold by measure for many years. TABLE NO. 9. CITIES IN WHICH METERS ARE EXTENSIVELY IN USE, Showing Amount of Reduction of Wastage and Number of Years in Which Results Were Obtained. PercentAverage REDUCTION OP Time in Percent- age of Daily WASTAGE. which age of ConYear Con- this re- total suners' CITY. Re- sump- duction Supply Supply ferred Per Gallons, was ac- Sold Sl to. Capita Per Per com- by by Gallons. Capita, centage. plished, Meas- MeasDaily. Years. ure. ue. Bamberg............ 1888 28.5........... 25 Bamberg............ 1895 15.3 13.2 46 7 44 Berlin............. 1868 26.1....... 36 Berlin........... 1880 16.6 9.5 36 12 78 Eisenach............ 1888 33.0...... 18 26 Eisenach............1894 12.4 20.6 62 6 50 99 Hamburg........... 1891 58.3...... 29 13 Hamburg........... 1899 47.0 11.3 20 8 49 21 Hannover........... 1890 34.0...... 1.. 9 20 Hannover........... 1894 15.8 18.2 54 4 68 100 Leipzig............ 1889 27.7....... 20 Leipzig............. 1891 15.3 12.4 45 2 90 Offenbach........... 1899 25.6..... 34 Offenbach........... 1900 17.1 8.5 33 1 53 Rudolstadt........ 1895 21.9.... 29 Rudolstadt.......... 1898 14.3 7.6 35 3. 100 Stettin............ 1894 31.0.... 13 Stettin............. 1896 20.0 11.0 33 2 76 Stralsund.......... 1893 40.2....... 48 Stralsund.......... 1896 21.4 18.8 47 3 78 189 2.4.1' /.7 In these cities it will be seen that in every instance there has been a substantial reduction of wastage, amounting to from 20 to 62 per cent of the total supply. Omitting Hamburg, in which only 21 per cent of the consumers' water is metered, the average reduction of wastage would represent a saving of about 43 per cent of the water supplied to these cities, the variation ranging from 33 to 62 per cent. 36 Part I: Methods of Controlling Loss. TABLE NO. 10. CITIES IN WHICH COMPARATIVELY FEW METERS ARE IN USE: Showing in Every Case an Irrational Fluctuation of the Rate of Consumption, with a Pronounced Tendency to Increase More than to Decrease. Average Daily Increase Reduction PercentConsump- In Gallons In Gallons Number age of CITY. Year. tion Per Capita, Per Capita, of Years Supply Per Capita, Per Per Included. Sold by er ______ Gallons.apita Day. Day. Measure. - Gallons Augsburg...... 1888 61.8.......... 3.... 1890 56.8.... 5.0 2 3.... 1895 63.8 7.0.... 5 1 Libeck....... 1869 14.5........... 1882 63.6 49.1.... 13 "1888 58.3...... 5.3 6.......1894 95.3 37.0.... 6 1896 79.2.... 16.1 2 7 1898 92.0 12.8..... 2 8 1900 82.8.... 9.2 2 9 Wfirzburg.... 1892 42.0........ 9.. 9 1893 49.0 7.0.... 1 8 1894 44.8.... 18944.2 1 8 1896 58.4 13.6... 2 12 1897 56.6.... 1.8 1 10 Regensburg... 1887 30.1.......... 3 1892 41.6 11.5.... 5 10 1897 34.9.... 6.7 5 18 Konigsberg... 1892 15.9......... 13 1894 20.1 4.2.... 2 16 1896 17.4.... 2.7 2 41 Kiel........ 1889 16.6.......... 12 ".......... 1890 24.0 7.4.... 1 11 1891 20.1........3.9 1 13 1895 25.9 5.8... 4 9 '......... 1896 21.9.... 4.0 1 10 Metz.......... 1894 25.9.......... 12......... 1897 34.6 8.7.... 3 12......... 1898 32.4.... 2.2 1 12......... 1900 35.7 3.3.... 2 12 Dessau........ 1890 19.5........... 14 1898 34.0 14.5.... 8 15 Karlsruhe..... 1880 20.8............ 25 1883 19.5.... 1.3 3 25 1888 25.3 5.8.... 5 25 1893 34.6 9.3.... 5 26 1894 30.6.... 1894.0 1 29.... 1895 37.0 6.4.... 1 25 1899 32.0.... 5.0 4 33 Bremen....... 1896 21.6.......... 25 "...... 1900 25.7 4.1.... 4 28 (Table No. io continued on next page.) 37 Waste of Water and Its Reduction. TABLE NO. 10.- Continued. Avege Increase Reduction PercentConsap- In Gallons In Gallons Number age of CITY. Year. tion Per Capita, Per Capita, of Years Supply Per Capita, Per Per Included. Sold by Gallons. Day. Day. Measure. Bernberg..... 1888 17.2.......... 14..... 1893 22.4 5.2.... 5 16.... 1895 6.3.... 6.1 2 10.... 1900 27.5 11.2.... 5 8 Danzig........ 1875 25.6.......... 6 "....... 1888 27.4 1.8.... 13 37....... 1890 25.1.... 2.3 2 45 "...... 1892 27.4 2.3.... 2 39 "....... 1895 23.1.... 4.3 3 44 Frankfort 0-0 1888 10.6.......... 43 1895 14.3 3.7.... 7 43 Miillhausen.... 1891 18.8.......... 57 "... 1898 37.2 18.4.... 7 59 "... 1900 27.0.... 10.2 2 60 Nuremburg.... 1889 16.1........ 52... 1892 20 8 4.7.... 3 56... 1897 20.3.... 0.5 4 63 Dresden....... 1876 12.4.......... 41...... 1898 26.7 14.3.... 22 76 The wide range of fluctuation in the rate of consumption in each of these unmetered cities, Table 10, is apparent and characteristic of cities where water is sold a discretion, or where the meters are practically confined to manufacturing and trades uses. It will be seen that in every city in this list, except Danzig, the total increase exceeds the total decrease by a large percentage. In Danzig apparently a sufficient proportion of the domestic consumption is metered to hold the total consumption down where fluctuations may be in either direction. 38 Part I: Methods of Controlling Loss. TABLE NO. I 1, CITIES IN WHICH PRACTICALLY ALL CONSUMERS HAVE BEEN METERED FOR MANY YEARS. CITY. Year. Average Daily Consumption per Capita, Gallons. Erfurt.......I ~18 90 1 11. 4 Breslau...... Colmar...... Chrd tnbri. Mhagdoenburg.. Potsdam... Qudinr. Strassburg. Weiar Wom........... HIdeleg 1896 1889 1896 1890 1894 1897 1900 1889 1893 1894 1896 1888 1889 1900 1890 1893 1894 1896 1898 1899 1885 1889 1896 1888 1899 1889 1899 1889 1900 1889 1900 1882 1893 1896 1878 1893 1896 1889 1900 1878 1897 1893 1900 1888 1899 13.5 19.5 23.2 14.8 21.1 16.1 18.8 9.5 15.9 14.5 16.1 18.7 13.0 21.4 15.0 9.8 12.7 9.0 14.0 10.3 27.2 23.2 24.8 5.8 7.6 4.2 22.7 13.7 22.5 5.8 17.2 4.5 18.5 12.7 4.8 10.0 8.2 14.5 27.2 11.3 14.5 22.2 28.8 21.9 21.4 Increase in Gallons Per Capita, per Day. 2. 1 3.7 6.3 2.7 6.4 1.6 8.4 2.9 5.0 1.6 1.8 18.5 8.8 11.4 14.0 5.2 12.7 3.2 6.6 Reduction in Gallons Per Capita, per Day. 5.0 1.4 5.7 5.2 3.7 4.3 4.0 1.8 0.5 86 No. of Years Included Percentage of Supply Sold by Measure. 6 4 3 3 4 1 2 1i 11 3 1 2 2 1 4 7 11 10 11 3 15 3 11 19 11 77 77 78 To all consumers To id 41 I$ It Id 88 79 80 85 75 83 88 all consumers 88 90 To all consumers 71 74 69 75 all consumers To 71 64 76 To all consumers 85 74 93 88 32 66 39 Waste of Water and Its Reduction. In the above cities, where practically all the consumers have been metered for several years, there is generally a tendency toward higher rates with increasing population; in Gotha there has been a material reduction; in Magdeburg a slight reduction. In nearly all the German cities of importance water was formerly sold at rates based upon a percentage of rentals, upon assessments, or upon fixture rates. In all these cities difficulties with wastage have been experienced and the amount of wastage has, in many cases, been relatively as great as in other countries where greater amounts of water are used. In many of these cities the methods used to control wastage have been the familiar ones of house-to-house inspections, of changing to the method of selling water by measure, and of partial metering combined with inspections. In those where inspections have been in force the method was found efficacious to a degree depending upon the frequency and thoroughness of the inspections; when inspections were stopped wastage would, in a few years, be as great or even greater than before inspections were undertaken. When meterage was resorted to wastage began to cease, but full effects in waste reduction were not derived until nearly all of the consumers were paying for their water by measure. Metering of manufacturing consumers has had little effect, as a rule, on total wastage, for the reason that but a small proportion of the water supplied by the Municipal plants in Germany is consumed by this class of takers, the greater percentage going to domestic and small trades consumers. Therefore, to control wastage, the domestic consumers have had to be reached. Where metering has been resorted to inspections from time to time were necessary to discover fraudulent uses, such as tapping the service pipes back of the meters; and where complete metering has been combined with occasional inspections wastage has permanently ceased. The sale of water by meter has not tended to discourage its use to a point below the liberal needs of the consumers. In all the cities which have been completely metered for many years the per capita rate is slowly increasing in accordance with the requirements of better living and greater needs. 40 Part I: Methods of Controlling Loss. The water used for public purposes is about the same amount per capita in German cities as in American cities of the same size and character, the difference in consumption being entirely in the uses for domestic manufacturing and trades purposes. The manufacturing use being, as a rule, insignificant, the German water works plants have to provide water only for domestic consumption and under conditions where there are relatively few fixtures per unit of population; and consequently where, it being less convenient to draw water, less water is used and less can be wasted. It is also to be seen, taking Berlin as an example, a city as large as Manhattan Borough of New York, that although nearly all consumers live in rented properties where water rates are collected from the owners of the properties, and not from the consumers, the system of sale by measure has been a natural transition from the old, illogical system of sale a discretion, entirely brought about by voluntary requests for meters when a few chosen landlords had given out the word that the system was more economical and more advantageous than the former plan of payments based on rentals. Of all the cities where meters have been introduced not one has ever gone back to the former methods; and in not one has there failed to be a healthy and steady increase in the percentage of water sold by meter. In some the rate of introduction has been more rapid than in others, but they one and all testify to the fairness, logicalness, and propriety of the system. It may be taken as axiomatic that when a change has been made from old customs to new ones, a change for the worse will not long be tolerated by any public,-German or American, and we have, in these few German cities, buying water exclusively by measure, (having changed from the antiquated system of charge on the assessment plan) a population greater than the combined populations of New York, Chicago, Philadelphia, Brooklyn, Boston and Baltimore, not only endorsing the method of sale-bymeasure by tolerance, but emphasizing the endorsement by rapidly extending it. The conditions in Berlin were very similar to those in New York and Philadelphia-large holdings of property by compara41 Waste of Water and Its Reduction. tively few owners, and all owners at the mercy of presumably wasteful tenants. But in Berlin, as in every other city which has had the experience, it was found that the tenants did not wilfully waste water. The wastage attributed to tenants was chargeable to careless landlords who, to personally save a few dollars in plumbers' bills, preferred to let the water run and have its value charged up in the taxes. Summing up the experiences of the foregoing American and German cities it may be stated: First.-Wastage and leakage can be quickly stopped and permanently suppressed without depriving any citizen of the use of as much water as that to which he has been accustomed. That the question of the suppression of the greater part of the wastage is to all intents and purposes merely a hunt for leaks followed by the watching of the different consumers and the insisting that they keep fixtures from leaking. Second.-That sale of water by measure has no tendency to restrict the use of water, but does notify the owner when the leakage from the faucets and other fixtures, is serious enough to require attention. That wastage from domestic premises is rare, but that leakage is prevalent. That tenants do not wilfully waste water, and that the blame for the present conditions in New York rests not upon the hundreds of thousands of users of water, but upon the relatively few owners who neglect to keep their plumbing in order as required by the City ordinances. That the responsibility for this state of affairs rests primarily with the system under which the water is sold. That the sale of water by meter does not work a hardship on owners of property rented to tenants; in practically all cities, from Berlin to Asbury Park, the water taxes are liens on the property, and yet, in no city, where the system has been adopted permanently, and where the rates have been properly adjusted, would either the water-works managers or the inhabitants of the city go back to the old system. The data given in the preceding pages relating to the experiences of the various American and German cities are only sum42 Part I: Methods of Controlling Loss. maries and in the brief form stated would not afford sufficient proof of the two foregoing propositions regarding waste and leak control. A perusal of the detailed experiences in each city referred to, descriptions of which will be found in Appendix D, will, however, be instructive, and point out several important facts that should be understood in considering the advisability of the application of any special methods for the control of leakage and wastage in New York. 48 PART II. THE UNACCOUNTED FOR WATER. (See, also, Appendix A.) A LL devices and methods resorted to for the measurement of flowing water are subject to errors which can not be avoided under practical working conditions; therefore the results of all such measurements are but approximate and must be accepted as such. Although reasonable accuracy can be obtained in experimental measurements the limits of error must be greatly extended when considering what can be practically depended upon in dealing with a large quantity like that supplied daily to New York City and which must be sold to consumers in hundreds of thousands of small driblets. Therefore in no large city is it ever possible to account for all the water which is delivered into the mains in the streets. The difference between the quantity supplied to the City and the quantity accounted for by use, that is, by commercial, domestic, public, and manufacturing consumption, I shall refer to hereafter as "The unaccounted-for water." Under the best practicable conditions, even if all the water were sold by meter measurement, it is not probable that over 75 per cent of the total amount supplied to New York City could actually be accounted for; the 25 per cent not accounted for representing leakage from mains and service pipes outside the houses, uncertainties in the measurement of the water entering the city distribution system, impracticability of keeping all water-meters in perfect order at the same time, under-registration of meters, particularly those which have become more or less worn in service, the losses due to the failure of meters to register the very slight flows caused by dribbling drafts when tanks are nearly filled, the surreptitious uses of water which can not always be detected and the scattered uses of unmetered water, for various purposes, and of which no account is kept. What would be considered a proper and reasonable allowance for unaccounted-for water in New York City must be a matter of judgment, based.on local conditions and on results which have 44 Part II: Unaccounted-For Water. been obtained in smaller American cities where it has been possible to make close estimates of the total supply, and of the water actually taken through meters by consumers. While it may seem unreasonable, to those who have not given this matter special study, to assume that not over three-fourths of the water could be accounted for in New York City under the most favorable conditions considerable light will be thrown on the matter by an examination of the data given in Table 12 and in Appendix A. TABLE NO. 12. COMPARATIVE STATISTICS OF SEVERAL AMERICAN AND GERMAN CITIES, Showing Particularly the Percentage of the Total Supply Not Accounted For. (By Meters on Consumers' Supply Pipes and Careful Estimates of the Unmetered Waters.) Figures Given, Relating to Consumption of Water, Are, Unless Otherwise Noted, Per Capita Per Diem. Water Taken by Con- Per sumers; Gallons per Water Water Total Cent per Capita per day. for Not Ac- per Cap- Kind of CITY. Year p Public Counted ita per g of ServiManu- Uses for. Day. g Supply. ices fac- Do- Total. Galls. Galls. Galls. X I Meturing. mestic tered. Brockton... 1898 3.4 14.4 17.8 2.4 10.9 31.1 35P'mpd 79... 1899 4.0 15.0 19.0 2.9 11.5 33.4 35...... 80...1900 4.5 15.0 19.5 2.9 10.8 33.2 32...... 81...1901 4.5 15.6 20.1 3.0* 10.1 33.2 30...... 83...1902 5.0 16.0 21.0 3.0* 12.0 36.0 33...... 90...1903 5.5 16.6 22.1 3.0* 12.0 37.1 32...... 90...1904 5.1 15.5 20.6 3.0* 13.3 36.9 26...... 91 Boston..... 1880 25.0 i30.0* 55.0 2.0 29.0 86.0 34 Gr'vty.....1892 30.0 30. 0* 60.0 3.0 32.0 95.0 34...... Cleveland... 1904 40.0 j26.0 66.010.0* 20.0 96.0 21 P'mpd 49 Englewood.. 1888.............. 52P'mpd 100 FallRiver.. 1899 7.0 15.0 22.0 5.7 8.5 36.2 23P'mpd 93. 1900....... 21.4 7.1 7.9 36.4 22...... 94. 1901....... 20.4 6.9 6.6 33.9 20...... 95..1902....... 23.4 8.3 8.7 40.5 2...... 95 Hackensack 1884....... 148. Ot 308. 0t 48 P 'mpd 100 1885....... 81.0t 238. Ot34......100 1886......... 101.0t256.0 t39......100 1887 131.0t 313.0t 42...... 100 1888(-Yr.)...... 173. 0t 446.0 39. 100 Hartford... 1904 3.0*30.0 33.0 5.0* 24.0 62.0 39Gr'vty 99 Harrisburg.. 1891 33.0 30.0* 63.0 5.0* 54.0 122.0 44 P'mpd |.1892 43.0 30.0* 73.0 5.0* 44.0 122.0 36........1893 42.0 30.0* 72.0 5.0* 44.0 121.0 36...... `E..1894 37.0 30.0* 67.0 5.0* 36.0 108.0 33...... (Table No. 12 continued on next page.) 45 Waste of Water and Its Reduction. TABLE NO. 12.-Continued. Water Taken by Con- Per smr;GlosprWater Water Total Cent. Capita per Day for Not Ac- per Cap- J*0 Kind of CITY Year _ a______ Public counted ita per R. of ServiManu- Uses, for, Day, 0 Supply. ces fac- IDo- -Total. Gals. Gals. Gals. 0.Me__ _ _ _ _ u In. m estic __ __ _ _ __ tered. Harrisburg.. Lawrence.. Milwaukee.. Ridgefield.. Madison.. Syracuse.. Taunton... Ware. Wellesley.. W. Orange.. Woonsocket Worcester.. Yonkers... Bamnberg... Berlin. 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 19021 1903 1904~ 1904 1886 1887 1888 1897 1898 1899 1900 1901 1902 1903i 1904 1904 1898 1899 1900 1901.1902 1903 1904 1902 1901 1902 1903 1904 1905 1902 1902 1900 1901 1902 1903 1904 1895 1896 1897 1895 1896 44.0 42.0 47.0 49.0 65.0 73.0 80.0 79.0 78.0 81.0 11.0 10.0 8.0 45.0 39.3 13.0 13.0 14.8 14. 15.2 16.6 14.7 0.3 0.4 0.4 16.0 22.0 22.0 24.0 30. 0* 30.-0* 30. 0* 130. 0* 130.0* 30. 0* 30. 0* 30.0O* 30. 0* 30. 0* 15.0 15.0 17.0 25. 0* 31.0 18.3 22.5 20.5 22.5 20.5 21.0 21.5 24.3 25.4 28.2 28.6 18.0 19.0 19.0 20.0 20.0 74.0 72.0 77.0 79.0O 95.0 103.0 110.0 109.0 108.0 111.0 26.0 25.0 25.0 70.0 14.0 14.0 14.0 15.0 20.0O 21.0 21.0 21.0 70.3 31.3 35.5 35.3 37. 1 35.7 37.6 36.2 24.6 25.7 28.6 29.0 40.0 39.6 46.0 49.0 51.5 9.6 9.2 9.8 15.3 15.0 5.0O* 5.0* 5.0* 5.0O* 5.0O* 5.0 *, 5.0O* 5.0O* 5.0* 5.0O* 5.0O* 5.0*i 5.0O* 5.0* 13.0O* 13.0O* 13.0O* 13.0O* 13.0O* 13.-0* 13.-0* 13.0O* 18.0 3.0* 3.0O* 3.0O* 3.0* 3.0O* 3.0O* 3.0O* 2.8 2.2 3.2 2.5 0.4 3.0 3.0 2.0 2.0 3.7 3.7 3.3 1.8 116 40.0 30.0 25.0 28.0 22.0 25.0 20.0 20.0 30.0 30.0 23.0 12.0 12.0 14.0 62. Of 10.01 -24. Of 17.0 13.0 15.0 15.0 17.0 12.0 22.0 37.0 20.0 14.3 15.2 23.7 24.9 15.3 13.4 24.8 17.0 20.6 22.1 23.2 23.5 7.0 29.0 37.6 41.4 40.0 37.0 40.5 2.0 2.4 2.7. 9 1.6 119.0 107.0 107.0 112.0 122.0 '133.-0 135. 0 1134.0 143.0 146.0 54.0 42.0 42.0 89.0 175.0Of 133. Of 181. Of 44.0 40.0 42.0 43.0 50.0 46.0 56.0 71.0 108.3 48.6 53.7 62.0 65.0 54.0 54.0 64.0 44.0 48.0 50.0 55.0 55.0 29.0 68.0 78.0 84.0 89.0 88.0 94.0 15.3 15.3 15.8 18.0 18.2 34 28 23 251 '9 '5 '5' 21 21 42 29 29) 35~ 8 12i 39~ 32 36 35 34 26 39 52 '9 129~ 281 i38, 38 28 25 39, 39 43' 441 42 20 24 ~42 48 49 451 421 43i '7 5 9 P.p P'mpd P'mpd P 'mpd b~r'Vt~y, P'mpd Pr'vty P 'mpd Pmpd ba 84 -100 100 =100 94 96 96 72 400 410 42 43 45 45 100 410 87 95 94 94 990 100 710 100 100 I I (Table No. 12 continued on next page.) 46 Part II: Unaccounted-For Water. TABLE 'NO. 12.-Continued. Water Taken by Con- W W per sumers; Gallons per Water Water Total s n Cent. Capita per Day. fo NotAc-perCap-.Kind of CITY. Year[ _____ __Public counted ita per K of ServiManu- Uses, for, Day, Supply. ces fac- Do- Total. Gals. Gals. Gals. 6 Me___turing. mestic tered. Eisenach... 1895......7 6.1 1.5 14.3 iiGr'vty 99 1896..... 8.4 5.3 1.5 15.2 io...... 99 1897... 8.3 4.9 1.1 14.3 9......99...1898.......8.2 3.7 1.3 13.2 io..... 99 "...118991...... 9.5 3.4 1.4 14.3 9...... 99 Hannover.. 1894.......10.8 2.8 2.2 15.8 14lP'mpd 100 d..11895......116.9 4.5 2.9 24.3 12...... 100 "..18961......15.8 2.0 3.4 21.2 i6...... 100 Rudolstadt. 1898..... 9.9 0.2 4.2 14.3 29 P'mpd 100.1899....10.4 0.3 4.4 15.1 29......100 Stralsund.. 1896.. 12.9 1.6 6.9 21.4 32...... 100 Breslau..... 1889.......13.5 2.5 3.5 19.5 iSP'mpd 100 "..../1890..... 14.0 2.6 3.7 20.3 i8...... 100 "....11891..... 13.9 2.4 3.8 20.1 19......100 "....11892.....13.9 2.8 5.0 21.7 23...... 100....1893......14.7 2.6 4.6 21.9 21..... 100....1894.......14.7 2.9 4.6 '22.2 21...... 100....11895......14.5 2.8 4.4 21.7 201,..... 100....1896....... 17.4 1.9 3.9 23.2 71...... 100 Darmstadt. 1890................ 100 1891.......11.2 0.7 1.6 13.5 i P'mpd 100 Id 1892....... 12.5 0.6 1.7 14.8 i6...... 100 1893..... 114.6 0.5 2.9 18.0 6......100 1894.... 16.0 0.6 3.2 19.8 211..... 100 I 1895......114.5 0.5 2.4 17.4 I3...... 100 1896......14.8 1.6 3.1 19.5 61...... 100 1897....... 15.2 1.6 3.5 20.3 171...... 100 1898...... 17.5 0.8 2.0 20.3 io...... 100 1899.......18.8 0.4 2.2 21.4 io...... 100 1900..... 18.8 0.4 2.2 21.4 io...... 100 Gotha.....1890.... 8.2 2.8 4.0 15.0 26...... 100. 1891.... 1... 7.4 2.5 3.3 13.2 25 Gr'vty 100 ".....18921.... 6.1 2.1 3.1 11.3 281..... 100.....11893...... 5.3 1.8 2.7 9.8 281..... 100.....11894...... 6.6 2.4 3.6 12.7 2901...... 100 "..... 1895.......6.6 2.4 3.6 12.7 30...... 100 ".....11896....... 5.3 1.2 2.5 9.0 281...... 100 "..... 18971.... 7.4 1.7 2.6 11.6 261..... 100.....]1898....... 8.4 2.1 3.5 14.0 251...... 100.....1899.......6.4 1.6 2.3 10.3 221....... 100 Magdeburg. 1891.......19.0 1.5 3.0 23.5 I31P'mpd 100.1892....... 19.4 1.6 2.8 23.8 ii...... 100.1893....... 20.2 1.8 3.6 25.6 i41...... 100 ".11894.......20.2 1.6 4.6 26.4 71.......100.1895.......18.3 1.2 3.5 23.0 51....... 100.18961.....19.7 1.3 3.8 24.8 I5...... 100 Mainz..... 1895.......7.41.9 1.0 10.a3 gP'mpd 100 ".....1896....... 8.7 2.6 0.6 11.9 5......I100 ".....1897....... 9.4 1.8 0.9 12.1 7...... 100 ".....1898....... 9.9 1.7 0.8 12.4 6....... 100.....1900 0.33.6 3.3 17.2 19.......100 (Table No. 12 continued on next page.) 47 Waste of Water and Its Reduction. TABLE NO. 12.-Continued. Water Taken by Con- Wat Wae T ter sumers; Gallons per oater Water Total Cent. Capita per Day foi NotAc- perCap- Jl Kind of CITY. Year Public counted ita per V of Servi - Manu- Uses, for, Day, VSupply. ces fac- Do- Total. Gals. Gals. Gals. st Meturing. mestic tered. MannheimAd it it Potsdam... Que dli nburg IV 41 94 Stratssburg. Wt7eimnar..., is 49 di di As Wo7rms.. 1892 1893 1894 1895 1896 1897 1898 1899 1900 1882 1892 1893 1894 1895 1896 1888 11889 [1891 1892 1893 1894 1895 1896 1896 1897 1898 1899 1900 1890 1891 1892 1893 1894 1895 1896 1897~ 1897 1898 1899 1900.. ~~~ r ~~ 12.2 13.7 12.3 13.5 14.2 '15.0 16.4 16.7 16.7 3.2 13.6 12.2 14.2 12.6 10.0 3.4 3.8 5.2 5.3 6.1 5.3 5.5 5.7 10.1 10.2 11.1 12.3 13.2 9.6 9.9 10.0 12.5 11.7 10.5 10.6 9.5 16.7 19.2 21.2 24.2 3.3 4.2 3.8 3.8 4.0 3.6 3.8 3.3 2.4 0.4 1.9 2.4 2.2 1.0 1.1 0.8 0.8 1.2 1.4 1.2 0.8 0.6 0.7 8.8 6.8 7.8 9.0 11.2 1.4 2.3 1.8 1.7 1.7 1.5 1.4 1.2 2.1 2.5 3.8 2.6 2.7 2.9 2.4 2.5 1.9 2.2 2.0 2.7 3.4 0.9 1.7 2.8 2.1 0.9 1.6 0.6 0.9 1.0 3.3 2.7 1.3 1.5 1.8 2.2 2.0 2.2 2.4 2.8 2.7 2.0 3.2 2.4 3.5 2.2 2.5 3.8 2.3 3.6 2.2 2.0 18.2 20.6 18.5 19.8 20.1 20.8 22.2 22.7 22.5 4.5 17.2 17.4 18.5 14.5 12.7 4.8 5.5 7.4 10.0 10.0 7.4 7.6 8.2 21.1 19.0 21.1 23.7 27.2 13.7 14.2 15.0 16.6 16.9 14.2 14.5 14.5 21. 1 25.3 27.2 28.8 15 14 113 I0 Io 9 12 '5 '9 TO II 6 12 12 i6 '4 33 27 '7 20 22 10 I0 I0 10 I0 20 '4 21 '5 21 '5 '7 26 II '4 8 7 P 'mpd P 'mpd P 'mpd 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 010 100 100 100 Data for the American cities taken from the Annual Reports of the Water Boards of the different cities. Data for the German cities from Die Stiiutische Wasserversorgung in Deutschen Reiche sowie in Einigen Nachbailindern. E. Grahn, 1902. * Figures starred (*) are estimates by 3. H. F., based on data given in reports, in most instances, or on such descriptions of conditions and uses as would warrant an approximate estimate. t Figures marked with a dagger refer to consumption per water taker; not per capita. Detailed information regarding conditions in the cities quoted will be found in description of waste reduction methods used in each, given in Appendix D. 48 Part II: Unaccounted-For Water. In practically all the foreign cities quoted in this table allowances have been made for slip of pumps and under-registration of meters. Without these allowances the unaccounted-for water in Berlin, for instance, would havy been from 20 to 25 per cent, which corresponds pretty closely with the best results obtained in this country. In these German cities, as in our smaller cities, the difficulties of keeping track of the water do not begin to compare with those to be expected in a city like New York. With these facts in mind it is my judgment that to account for three-fourths of the water would represent about the highest limit attainable in New York, and for conditions as they are likely to exist here, with changes of administration, withholding of appropriations, and other practical questions to be contended with, it is more than likely that not over 70 per cent could be accounted for year after year, even with complete metering. This does not mean that 30 per cent of the water would be wasted or lost by leakage; but rather, that this extra quantity of water would have to be furnished in excess of that which the consumers' meters would account for, after making proper allowances for all the water used for public purposes, to balance the errors and losses that can not be prevented. (See Appendix A.) 40 PART III. NON-DOMESTIC USES OF WATER. THE water used for manufacturing, trades, and commercial purposes in New York is largely sold by meter, over half the total revenues of the Department, for Manhattan and the Bronx, being derived from the metered water; and as all the metered water is sold at the uniform rate of 10 cents per 100 cu. ft., the revenues of the Department from this source afford data for an estimate of the quantity of water used by the nondomestic consumers. Commissioner Dalton, in the Annual Report of the Department of Water Supply for 1898, stated that in that year the 35,442 meters in use in Manhattan and the Bronx covered practically every place where water was used to any considerable extent for other than domestic purposes. In view of recent statements made by the Water Registrar it is estimated that there may be at the present time approximately 10,000 services, other than domestic, upon which meters should yet be placed. Besides the water sold by meter at the docks for steamboat uses there is a considerable amount taken by tugs at the water front and by contractors, for building purposes, supplying boilers, and other miscellaneous uses, which is not metered, and in recent years has provided approximately about two-thirds as much revenue as the metered water sold for shipping. There is no way by which the quantity thus sold can be estimated correctly, but even if calculated at regular meter rates its amount would average about 5 per cent of the total metered supply. Undoubtedly the amount actually taken for these purposes is considerably in excess of this estimate, as an investigation by Water Registrar Padden in March of the present year brought to light that tugs holding permits to take water from the hydrants along the River front at a stated sum per year, without regard to quantity, had abused the privilege to the extent of taking water at one hydrant, selling it to some vessel, and then watering up again at some other hydrant. This has been stopped by assigning 50 Part III: Non-Domestic Uses. certain hydrants at which certain tugs shall get their supply of water, and taking other official steps to prevent trafficking at the City's expense. How much water has been stolen in this way can not be estimated; but when taken in conjunction with the fact that there had been discovered, in the early part of the year, many large unmetered buildings in the downtown section and that it is estimated that at the present time the number of meters actually in use should be increased about 25 per cent to cover all except domestic services it is plain that the manufacturing and commercial uses as calculated from the sales of metered water must be too small by a considerable percentage. If these 10,000 properties should consume water at the same average rate as those fully metered the correction would amount to 25 per cent of the reported metered consumption; this, however, would be too large an allowance as the premises now metered include all the large users, whereas those still requiring metering are mostly relatively small consumers. As a provisional figure to arrive at a reasonable estimate of the total use of water apart from that required for domestic and public purposes, and which would be classed as water used for manufacturing, trades, commercial, building and shipping purposes, I have assumed that the extra use by these 10,000 unmetered consumers would amount to about 10 per cent of the reported metered water; and have therefore increased the reported metered consumption by 15 per cent, 10 per cent to cover the buildings not metered and 5 per cent to cover the water taken for shipping, building and special uses now paid for at special rates. In view of the many opportunities for the use of water without limit, particularly by contractors and others under yearly permits, and of a probability that a pretty large amount of water is being used without payment therefor being made to the City, the allowances above stated are not too large. On this basis, the estimated amounts of water used per capita of the total population in Manhattan and the Bronx for manufacturing, commercial, building, trades and shipping purposes, in other words, all the water used except that taken for domestic and public purposes, would be as shown in Table No. 13. 11 TABLE 13. SHOWING QUANTITY OF WATER SOLD BY METER MEASUREMENT IN MANHATTAN AND THE BRONX, as deduced from the cash sales. All metered water is sold at the rate of Io cents per Ioo cubic feet, or $I33.33 per million gallons. Estimated amounts used for manufacturing, commercial, trades, building, shipping and other non-domestic uses obtained by increasing amounts of metered water 15 per cent.; of this allowance 5 per cent. is presumed to cover unmetered water used for shipping, building purposes, boilers and miscellaneous non-metered uses, and Io per cent. the uses in properties remaining unmetered but which should be classed as commercial trades and manufacturing. In the premises now metered are included several hotels, apartment houses, clubs and domestic properties of the better class. No reduction has been made for these uses in estimatin the probable amount of water used for non-domestic purposes for the reason that it is believed the amount of water now taken surreptitiously from the city, and which should be paid for by meter, will more than offset the quantity used in these hotels, apartment houses, clubs and residences. me CORRESPONDING QUANTITIES Probable OF WATER. Probable Average Actual _ _ Average Daily Use Receipts from Actual Amounts Total Sales Daily Use for NonMetered Water Receipts from Returned in Metered Water. Estimated of Metered Domestic YEAR Exclusive of Dock and Arrears. Manhattan Population Water. Purposes. Riverdale and Steamboat Metered and Cubic Feet Million Supplied Gallons Gallons Steamboat Meters. Supply. The Bronx. per Year. Gallons per Capita per Capita meters. per Day. D per Day (Col.9+ 15%) 1894 $1,659,110 $99,660 $263,542 $2,022,312 2,022,000,000 41.5 1,850,000 22.5 26.0 1895 1,706,690 100,000* 264,414 2,071,104* 2,071,000,000 42.6 1,869,000 22.7 26.0 1896 1,891,480 105,331 296,759 2,293,570 2,293,000,000 46.8 1,890,000 24.8 28.5 1897 2,012,118 85,825 244,523 2,342,460 2,342,000,000 48.2 1,910,000 25.2 29. 0 1898 2,070,514 109,454 207,790 2,387,762 2,387,000,000 49.2 1,950,000 25.2 29.0 1899 2,200,000* 130,000* 220,000* 2,550,000* 2,550,000,000 52.0 2,000,000 26.0 30.0 1900 2,305,985 107,308 325,390 2,738,683 2,738,000,000 56.4 2,053,000 27.5 31.6 1901 2,455,446 130,971 323,896 2,910,313 2,910,000,000 60.0 2,100,000 28.5 33.0 1902 2,980,930** 151,663 200,000* 3,332,593* 3,332,000,000 68.5 2,160,000 31.7 36.4 1903 3,098,117** 138,136 200,000* 3,436,253* 3,436,000,000 71.0 2,220,000 32.0 37.0 1904 3,200,000t 165,000t 200,000t 3,565,000t 3,565,000,000 75.0 2,290,000 33.0 37.7 *About. **Total revenue from metered water, excluding steamboat meters. tEstimated from published reports for first three quarters of year. I., Q C-. C., 0 It Q I Q. N__4 &n f% Q. rr rl (Z Part III: Non-Domestic Uses. It has been impossible to obtain confirmation of the accuracy of the estimated quantities and amounts, relating to the metered water, from the office of the Water Registrar in time to incorporate the official figures in the foregoing tabulation. It is believed, however, that such revision as might be necessary would not alter the figures given in the last column more than one or two gallons per capita per day and that the figures as given are sufficiently accurate for present purposes in view of the necessity, in any event, of estimating the amounts of the unmetered quantities used for non-domestic purposes, which quantities are sold under yearly permits or special arrangements. 53 PART IV. WATER REQUIRED FOR DOMESTIC AND PUBLIC USES. THE actual amount of water required for domestic consumption in Manhattan and the Bronx varies between wide limits in different sections of the City; and in arriving at a reasonable estimate of the average amount necessary, the relative percentages of the different classes of consumers must be considered carefully. In certain completely metered New England cities measurements have been kept of the domestic consumption for several years. Of these cities the statistics of Fall River and Lawrence are both frequently quoted as indicating a proper per capita allowance of water for domestic purposes. These data, however, are misleading unless it is understood that in both cities a large proportion of the population is housed in tenements holding several families and having very much poorer water facilities than are to be found in average tenement districts in New York. The average domestic consumption of the two cities mentioned is about 15 gallons per capita per day. In Mr. N. S. Hill's tests, in 1903, of the consumption of water in District No. 2, a typical, crowded, East Side tenement section lhaving a resident population of 38,906 it was found that after deducting the plumbing leakage the average consumption per capita was 27.6 gallons per day. In a tenement district with a population of 8,000 in Brooklyn, Mr. De Varona found an average daily per capita consumption of 39 gallons after deducting from the total consumption the measured house leakage and the water supplied through meters, or 42 gallons per day, including the metered water. In District No. 5, in New York, which had a population of 32,200, and represented a residential apartment house district in the Bronx the average consumption, deducting house leakage, was 43.5 gallons per capita per day, and deducting the metered consumption, about 41 gallons. In a typical upper West Side apartiment house district with a population of 8,872 the consumption averaged from 140 to 170 gallons per capita per day on two dif54 Part IV: Domestic and Public Uses. ferent tests. In District No. 8, which included the Southern end of New York below Fulton street the consumption, based on the resident population in that district, was 860 gallons per capita daily, and based on the total population, resident and day combined, 83 gallons. A provisional estimate of the amount required for domestic use can be approximated by the aid of Mr. Hills' tests which included districts covering about 12 per cent of the area of Manhattan Island. The total amount of water supplied to these districts, excluding District No. 8, which is in the downtown office building section where practically all the consumption is metered, was 41,680,000 gallons per day. The total amount of fixture leakage in all the districts was about 6,100,000 gallons per day, or approximately 15 per cent of the total supply. The population resident in the districts tested was nearly 500,000 or about 25 per cent of the total resident population of Manhattan and the Bronx. 55 Waste of Water and Its Reduction. TABLE NO. 14. THE CONSUMPTION OF WATER FOR SEVERAL YEARS IN MANHATTAN AND THE BRONX, EXCLUSIVE OF THAT PORTION OF THE BRONX SUPPLIED BY YONKERS AND THE NEW YORK AND WESTCHESTER WATER COMPANY,t HAS BEEN AS FOLLOWS: YEAR. 1894...... 1895...... 1896...... 1897...... 1898...... 1899...... 1900...... 1901*..... 1902...... 1903t..... 1904...... POPULATION. 1,850,000 1,869.000 1,890,000 1,910,000 1,950,000 2,000,000 2,053,000 2,100,000 2,160,000 2,220,000 2,290,000 Average Quantity Supplied from the Croton Watex Shed, -Million Gallons per Day. 163 166 186 197 207 226 249 256 259 266 290 Average Quantity Supplied from the Bronx and Byram Watei Sheds-Million Gallons per Day. 13 14 15 15 19 20 18 11 18 18 19 Total SupplyMillion Gallons per Day. 176 180 201 212 226 246 267 267 277 284 309 Average Daily Consumption per capita, Gallons. 95 96 106 111 116 123 130 127 128 128 135 --- *Nov, 27, 1901, drafts on Byram River, amounting to 7,000,000 gallons per day discontinued in obedience to injunction issued by U. S. Circuit Court. Deficit made up by pumping from old aqueduct, which began in 1900, at the rate of 1,800,000 gallons per day, increasing to 10,000,000 gallons per day till July, 1902. This rate was maintained until October 15th of that year, when a 48" main was laid to connect the aqueduct directly with the street main, since which time the central and southern portions of the Bronx have been supplied with the same average amount of water from the Croton system without pumping. tWorks of N. Y. and Westchester Water Co., purchased early in 1903. 56 Part IV: Domestic and Public Uses. The total average amount of water supplied to Manhattan and the Bronx in 1903 was 284,000,000 gallons per day; Of this, 25 per cent, or 71,000,000 gallons could not be accounted for under the best obtainable conditions. 15 " " " 43,0000000 gallons would represent leakage from fixtures, calculated at the rate found in the test districts, and assuming the same average conditions over the entire city. 5 gallons per capita, or a total of..........11,000,000 gallons would represent the amount used in public buildings and for other public purposes. 37 gallons per capita or a total of.........82,000,000 gallons would represent the uses for manufacturing and commercial purposes. Total...............217,0000,000 Total supplied to City....254,000,000 gallons. Leaving for the purely domestic use of the entire City........67,000,000 gallons or 30.2 gallons per capita per day if all leakage and unnecessary wastage could be stopped. I believe this to be an approximately correct estimate of the average amount of water actually required for domestic purposes in Manhattan and the Bronx if house leakage were prevented. After years of careful investigation and study of the problem in Boston Mr. Dexter Brackett arrived at the same figure as a reasonable allowance for average conditions. Of course there will be places in New York where the average use will not exceed 8 to 10 gallons, while there will be others where perhaps as much as 57 Waste of Water and Its Reduction. 200 gallons per capita will be used and will be cheerfully paid for if sold by meter measurement. The reasonableness of the figure arrived at for average conditions, even though a large proportion of the inhabitants reside in apartments and tenements, will be conceded when it is remembered that perhaps in no large city in the world is the proportion of fixtures to population as great as in New York, particularly in the more modern buildings which have replaced old ones at a rapid rate during the last decade. This fact, in my judgment, would tend to make the average conditions for domestic consumption in New York more similar to those in Boston, Cleveland, Harrisburg, Milwaukee, Syracuse and Wellesley than to those in Brockton, Fall River, Lawrence, Providence, Taunton, and Yonkers. In view of such facts as can be applied to local conditions, I therefore regard an average allowance of 30 gallons per capita per day as the lowest that should be safely estimated for domestic consumption under present conditions in Manhattan and the Bronx, with all house leakage stopped. No complete local data are obtainable as to all the water used for public purposes, such as schools, public buildings, charitable institutions receiving free water, street washing and sprinkling, sewer flushing, blowing out mains, supplying water for fires, public comfort stations, parks, public fountains, and other necessary and legitimate purposes. The uses for these purposes vary by a rather large percentage in different cities, according to the character of the city and the extent of the uses for different purposes. In New York probably all the water devoted to public purposes might not average more than 2 to 3 gallons per capita per day as estimated by Mr. Brackett for the City of Boston; but in view of the variability exhibited in different years in different cities where careful estimates have been made of the amounts used each year, such as in Fall River, Wellesley, Yonkers and the German cities, I am confident that an average allowance of 5 gallons per capita per day will be about as close an estimate as can be made with safety from the available data. The total amount of water thus used, whether it be 2 gallons or 5 gallons per capita, is so small in comparison with the total amount supplied to the city that a considerable relative error in the allowance made therefor is really not of much significance. 58 Part IV: Domestic and Public Uses. The large amounts used for public purposes in Madison, Wis., and Syracuse, N. Y., (see Table 12) are due, in the case of Madison, to the enormous quantities used for sprinkling macadam streets, and other special uses, and in Syracuse, largely to wastage through public watering troughs and fountains. 59 PART V. LEAKAGE AND WASTAGE. THE controllable leakage and wastage in New York City, that is, the portion which can be effectively controlled in a reasonable length of time so as to permit a material reduction in the amount of the total supply, is practically limited to the unmetered consumption; in other words, to the domestic consumption. The amount of water used for public purposes, for fires, street sprinkling, sewer flushing, fountains, public buildings, and other public uses is, from the best data obtainable, and from a comparison with the amounts actually used in other cities, an insignificant proportion of the total supply, and, as previously estimated, probably not over 5 gallons per capita per day. It is not possible from any data obtainable from large cities in the United States to separate the losses due to leaks in the mains and service pipes from the other unaccounted-for waters. The uncertainties regarding the proper allowances for under-registration of meters and slip of pumps, or gauging of aqueduct flows, make any estimate of these amounts liable to error in so large a proportion of the total amount of unaccounted-for water that but lttle confidence could be placed in estimates of main leakage deduced therefrom. In a small supply estimates of this nature can and have been made; such estimates, however, would not apply to conditions in a large city like New York where mains are likely to break in some part of the City almost daily; and where, in years past, many old, leaky mains and services have been abandoned. While there is undoubtedly considerable leakage from the mains and service pipes in the streets there is every indication that the amount of this leakage is not excessive, or larger than in many smaller cities where it is possible to look after such defects much more readily than in a large city like New York. Such leakage as exists is probably scattered over the entire City and not much more prevalent in one section than in another; but as no detailed information is available on this point and as none of value could be secured short of many years of systematic and costly work in testing the flow in the mains speculation on either its location or exGO Part V: Leakage and Wastage. tent would be useless. The fact remains that even if it exists the multitudinous places that would first have to be found and dug up before the leaks could be stopped would keep a large and expensive force of inspectors and laborers busy for many years before enough leakage could be stopped to materially affect the total consumption of water in the City. (See Appendix C.) It remains, therefore, to point out that, as previously stated, the only leakage and wastage that can be quickly controlled is that which takes place within the buildings in the City; and as practically all business, manufacturing, and trades consumption is now metered, there is left for consideration only the wastage by domestic consumers. On the metered consumption the only improvement to be looked for will be in the direction of increased revenues if meters are looked after and tested more frequently than has been the case in the past, and the securing of increased numbers of metered consumers by the discovery of unauthorized or unknown connections with the mains in the streets or with metered services back of the meters. (See Appendix B.) While, manifestly, the amount of leakage and wastage will be different each year and can not, from any data now existing, be determined with exactness the approximate amount thereof in Manhattan and the Bronx can be deduced within reasonable limits of accuracy for several years past. An estimate of the average amount of preventable leakage and wastage in Manhattan and the Bronx may be arrived at in the manner shown in Table 15, in which the figures for the rate of consumption are given in gallons per capita per day. 61 Waste of Water and Its Reduction. TABLE NO. 15. YEAR. DISTRIBUTION OF — ____ — SUPPLY. 1894 1895 1896 1897 1898 1899 1900 19 01 1902 190 3 1904 Domestic use alone, average.. 30 30 30 30 30 30 30 30 30 30 30 Public uses, average........... 5 5 5 5 5 5 5 5 5 5 5 Office buildings, c o m m e r cial, manufacturing, shipping, building, etc..... 25 26 28 29 29 30 32 33 36 37 38 Total use........ 60 61 63 64 64 65 67 68 71 72 73 Total supply..... 95 96 106 111 116 123 130 127 128 128 135 Difference not accounted for.... 35 35 43 47 52 58 63 59 57 56 62 Percentage of total supply not accounted for. 37 36 40 42 45 47 48 46 44 43 46 Per cent. Average percentage of total supply not accounted for by use during whole period........................................... 43 Reasonable percentage of unaccounted-for water, assuming the City to have been completely metered for several years....... 25 Probable least percentage to which the unaccounted-for water can be reduced before new supply can be introduced, even with universal m etering......................................... 28 Difference, representing the percentage of the total supply that may probably be saved by stopping all house leakage......... 15 G2 Part V: Leakage and Wastage. The placing of meters alone will not stop all the leakage in the buildings; small and seemingly inconsequent leaks will always be more or less prevalent in the houses in a city as large as New York, and even though this leakage will be evident in the water bills many owners will prefer to pay the small additional price for the wasted water rather than spend any considerable sum for repair work, until the leaks become large enough to make the payments actually burdensome. I should estimate, as a result of general observation, that complete metering would permanently control in New York an average of all but one or two gallons per capita of house leakage; and that it would take at least five years of active, aggressive work in the placing of meters to secure this result. Even at this rate there would have to be placed an average of about 14,000 meters per year to secure in five years all the benefits possible. I know of no city where such a large number of meters has been set in any year, about 4,000 being the record to date, so far as I know, which was approached in Newark, Hartford and Milwaukee; there is no reason, however, to believe it impracticable to set 14,000 per year with a large organization under efficient management. If the placing of meters is begun in the tenement and apartment house districts, where the population is densest and the proportionate leakage per person greatest, the first year's work will probably result in stopping about 35 per cent, the second year 60 per cent, the third year 75 per cent, the fourth year 90 per cent, and the fifth year all the controllable house leakage. These figures are estimates based on the results that have been obtained in other cities and will only be reached by an aggressive campaign coupled with inspections of each district metered, and warnings to all owners that plumbing fixtures must be put in repair immediately and before the meters are set. The results obtained by the first three or four years' work will not be of permanent benefit, however, unless before stopping, the meter system be extended to practically all consumers and the further policy be adopted of metering all new supplies as fast as connections are made with the city mains. Under this arrangement the question of leakage and wastage will have been settled for all time. 63 Waste of Water and Its Reduction. it will be noted that no mention has been made of the effect of the large day and transient population on the domestic consumption in Manhattan. Nearly half a million people resident on Long Island and in suburban towns in New Jersey, New York and Connecticut engaged in various commercial, professional, and manufacturing pursuits in New York City spend their day hours at their places of business and return to their homes at night. This large population, however, is all taken care of in the metered consumption at the present time as all places not residences are presumed to be metered. For this reason it is not necessary to secure detailed data regarding the rate of increase of this population, or the probable increase of consumption due to its presence, for the purpose of estimating the amount required for domestic consumption alone. PART VI. ON THE AMOUNT OF WATER THAT MAY BE HAD FROM THE CROTON WATER-SHED. T is very essential that the capacity of the Croton water-shed to furnish water should be clearly understood in order that the gravity of the present situation may be fully appreciated. The natural flow of the Croton River varies from day to day, the greatest monthly flow recorded having occurred in January, 1874, during which month the average yield of the river at the old Croton Dam was at the rate of 1,585,000,000 gallons per day, while the least average monthly flow, only about 10,000,000 gallons per day, occurred during September, 1881, and September, 1883. In every year there are months of heavy flow and months of very low flow. This variability depends not so much on the amount of rain falling during the year as by the way in which it is distributed through the different seasons. These influences are too complex for precise analysis. Generally speaking, heavy rains during the Spring combined with an early disappearance of frost will produce well-sustained summer flows even though the summer rains be much below the average. From Table 16 it will be seen that in some years the flow of Croton River at the old dam has been nearly 100 per cent greater than in others having about the same rain-fall. 65 Waste of Water and Its Reduction. TABLE NO. 16. FLOW OF CROTON RIVER. on the ater Shed Average Natural Flow of Croton YEARDuring the Entire River for entire year. GalDuring the Entire Ions per Day. 1874.................. 42.37 405,000,000 1883.............. 42.70 215,000,000 1875................. 40.68 339,000,000 1872.............. 40.74 272,000,000 1895.............. 40.58 257,000,000 1873.................. 43.87 404,000,000 1885.............. 43.67 286,000,000 1868.............. 50.33 536,000,000 1884................. 51.28 387,000,000 1889.................. 55.70 506,000,000 1878.............. 55.70 438,000,000 Another extreme condition is exhibited by comparing the average daily flows for 1874 and 1897. Inches of rain Falling on the Water Shed Average Natural Flow of Croto,; YEAR. During the Entire River for Entire Year. GalYear- Ions per Day. 1874................. 42.37 405,000,000 1897.................. 53.12 413,000,000 Thus, in 1897, with 25 per cent more rain than in 1874 the average flow of the river was only increased 2 per cent. These complex conditions from which no precise laws can be deduced, since they depend on the distribution of the rain-fall, on its total amount, on its amount at times when the ground is full of water from previous rains, on the amount of water taken up by vegetation, on'the prevailing conditions governing evaporation, such as the winds, the capacity of the air to take up moisture, etc., make it impossible to calculate or to estimate from the rain-fall records, with any reasonable degree of accuracy, what the flow of the stream has been in the past or is going to be at any particular time in the future. It is possible, however, from the actual records of the flows in past years, to state with considerable accuracy what probable least and average flows are likely to be expected in the future and what the average flow has been for about 38 years past. 66 Part VI: Capacity of Croton Water-Shed. When storage reservoirs of considerable capacity are built on a stream two elements are introduced which modify the stream flow: First.-Evaporation and leakage through pervious strata underlying the reservoir site by which some of the stream flow is positively lost. Second.-When stored in natural reservoirs the water slowly soaks into the soil surrounding the reservoir and continues to do so until an impervious stratum is reached. When a reservoir where conditions are favorable for such ground storage is drawn down, therefore, a part of the water which has soaked into the soil will again run back into the reservoir, increasing the quantity obtainable beyond that which was actually visible in the reservoir. These two influences, therefore, tend to operate in opposite directions, one to decrease the yield of the stream and the other to increase it. In the Croton water-shed the opportunities for ground storage are not favorable and the net result of storage is to decrease natural stream-flow by practically the amount lost by evaporation and leakage. The larger the reservoir, therefore, that is, the greater the relative water surface exposed to the air, the greater the evaporation and the greater the loss of water. It will be seen therefore, that if a reservoir had a large enough exposed surface, and no water at all were drawn out, it might be possible for the entire flow of the streams entering it to be lost by evaporation. This is precisely the condition that obtains at the Dead Sea, the Great Salt Lake, and the Oceans; evaporation equals rain-fall and there is neither a gain nor a loss when considering a long series of years. Of course all bodies of water situated thus become salty in the course of time. The building of reservoirs tends to conserve for future use the heavy flows due to spring floods while, at the same time, the very existence of these reservoirs tends also to reduce the average flow of a stream in an amount equal to the evaporation from the water surface combined with what is lost by leakage through the bottoms of the reservoirs. It will be seen, therefore, that there is a 67 Waste of Water and Its Reduction. limit beyond which it is not possible to increase the average amount of water that can be drawn from a stream by means of storage reservoirs. This limit, for the Croton, and for the Sudbury River, which forms a part of Boston's water supply, is known from the records of past years and is found to be, in both cases, about 500,000,000 gallons for every square mile of the water-shed. With this amount of storage it would be possible to draw from the Croton River year after year, 360,000,000 gallons per day, which would be the average flow of the River as found from the records for the past 35 years when corrected for evaporation and other losses. Manifestly, no more than this could be continuously drawn under any possible conditions and therefore 360,000,000 gallons per day is the greatest amount of water that could be depended upon, from the Croton River, no matter how much storage were provided. To obtain this quantity, the storage reservoirs would have to hold an available supply of about 175,000,000,000 gallons, or nearly enough to supply the draft for two years without the help of stream flow and more than double the amount of storage now available; and if the works had been built for these conditions in 1870, and this rate of draft had been maintained continuously thereafter, by the Fall of 1884 the reservoirs would have been empty, and would have remained so until the Spring of 1887, from which time until the middle of 1899 they would have slowly filled up again. This sets the limit as to the greatest possible amount of water that can be had from the Croton. Now as to the greatest amount that it is practicable to draw. The total storage capacity of the present reservoirs, with the proposed Cross River reservoir added in, and including the water that can be stored in Central Park and Jerome Park reservoirs, special ponds not owned by the City, but'from which water can be drawn by arrangement, when necessary, and by using flashboards on the Croton dam, is about 82,000,000,000 gallons, or about half the total storage required to enable the water-shed to yield 360,000,000 gallons daily. The amount of storage already provided for, as above, however, is sufficient to permit an average daily draft of 850,000 gallons per square mile of water-shed, 68 Part VI: Capacity of Croton Water-Shed. or, 308,000,000 gallons per day, according to the experience on the Croton, or about 285,000,000 gallons per day according to the experience on the Sudbury; in round numbers, say, 300,000,000 gallons per day. There are two things to be considered in storing water for the. use of a city: quantity and quality. The limit of total quantity for the Croton has already been pointed out; the limit of permissible draft is governed not by the ultimate yield of the water-shed, but by the effect of storage on the quality of the water. Unlimited storage with no water drawn from the reservoir will ultimately produce salt water. A storage sufficient to enable the average flow of the stream to be taken daily from the reservoirs will ultimately, in artificial reservoirs, produce a water differing in no essential organic constituents from sewage. Such a reservoir in the Croton water-shed, operated as described, would permit no overflow for thirty years; and during this time the thousands of tons of organic matter which would be washed into it from the surrounding hills would decay, decompose, and be changed into complex chemical compounds serving as food for microscopic and minute plant forms which would multiply to such an extent as to shortly render the water unfit for use either for drinking or for steam purposes. With no storage, of course, only the daily flow of the stream could be depended on, which, as already stated, has averaged as low as 10,000,000 gallons per day and as high as 1,585,000,000 gallons per day for a month at a time. The practicable limit of storage to guard against undue deterioration of the water, therefore, lies between these limits and can be quite definitely fixed by the experience of the past gained not only on the Croton water-shed but on others in neighboring States. It has been found that whenever the storage has been carried beyond the point sufficient to permit a daily draft averaging much over 750,000 gallons per square mile of water-shed the water is 89 Waste of Water and Its Reduction. likely to begin to show signs of deterioration. With this amount of storage the reservoirs are likely to be held down below high water mark for two to three successive years at a time, owing to unfavorable rain-fall conditions, and in this time sufficient stagnation will take place, coupled with the growth of weeds, pondlilies, rushes, etc., in the shallow portions of reservoirs, to produce the discoloration of the water and the growth of numerous small forms of plant life which by their decay or by the secretion of minute globules of vegetal oils give rise to disagreeable or foul odors and nauseating tastes. A yield of 750,000 gallons per square mile per day from the Croton water-shed would correspond with an average draft of 270,000,000 gallons per day. The consumption by the city is now over 300,000,000 gallons per day. Any considerable development in excess of this limit will be at the risk of spoiling the water. As the reservoirs now in use, and provided for in the future, will yield about 20,000,000 gallons per day more than this safe limit it can readily be seen that trouble may even now be expected and that efforts to increase the storage in the Croton basin by any considerable amount in order to avoid the construction of the new Catskill supply should on this score be discountenanced. The limit given for the ultimate amount possible to be derived from the Croton water-shed, even with storage reservoirs twice as large as are now in existence, 360,000,000 gallons per day, represents the largest quantity that could be continuously derived from this source for a long series of years. During the present year, if the rate of consumption continues to increase in the ratio that has prevailed for the last ten years, the draft will be within about 3 per cent of this limit by December, 1906. In the Winter of 1895 and 1896 a draft 15 per cent heavier for the year than was actually made would have emptied the reservoirs. In 1901 an increase of 33 per cent would have emptied the reservoirs. The rate is at the present time about 5 per cent above what would be a safe limit if the Cross River reservoir, not yet under construction, were actually in service. Since about 1901 the rate of flow of the Croton river has been above the average; it must therefore soon fall below, and when it does the City will suffer; for with the demand above and the sup70 Part VI: Capacity io Croton Water-Shed. ply less than the average flow of the river there can be but one result-namely, the emptying of the reservoirs. This drop will probably come within the next three or four years if the behavior of the river as regards flow does not change entirely in character. There have been four such cycles since 1870, the first reaching its lowest point in 1872, recovering in 1876; the second reaching its lowest point in 1877, recovering in 1879; the third reaching its lowest point in 1885, recovering in 1891, and' the fourth reaching its lowest point in 1895 and recovering in 1899. There is no way by which to tell what will happen in the future but with the present draft so near the average flow of the river it is alarming to think what would result if a year or two of unfavorably distributed rain-fall should occur before the new works can be made ready. Unless leakage and wastage were under control it would mean a water famine. 71 PART VII. FUTURE CONDITIONS. T the present time it is probable that the amount of water actually used for domestic purposes in Manhattan and the Bronx averages about 30 gallons per capita per day. With the extensive development now taking place in the Bronx as a result of rapid-transit facilities, and the erection of detached dwellings as well as of moderate and high-class apartments almost without number in that district, combined with the rebuilding of many districts in Harlem which have heretofore developed slowly, it is certain that the net effect of these changes will tend toward increasing the per capita consumption for domestic purposes, for the reason that the better the classes of houses the greater the uses of water. In order to take this into account, a fact which has been observed in all American cities, I have assumed that in 1910 and from that time forward the per capita consumption will average about 35 gallons per day for the entire City, instead of 30 gallons. Similarly, with regard to the non-domestic uses, I have assumed an increase to 40 gallons per capita by 1910, with a continuance of that rate for subsequent years. In the premises now metered are included several hotels, apartment houses and clubs, as well as a few domestic properties of the better class. No reduction has been made for these uses in estimating the probable amount of water used for non-domestic purposes for the reason that it is confidently believed the amount of water now stolen from the City, and which should be paid for by meter, will more than offset the quantity used in these hotels, apartment houses and clubs. The effect of these assumptions, which I consider to be conservative, when viewed by past experience in New York, and by the conditions which have been found in other cities, are exhibited graphically on Diagram No. 1. Upon this diagram are drawn lines representing the estimated amounts of water, in million gallons per day, used for different purposes each year from 1894 to 1904 inclusive, with estimates, based on the foregoing assumptions, for these same uses for several years in the future. There are also placed upon the diagram 72 0 0 N 0?O** 0 Q) r Ncmo Part VII: Future Conditions. horizontal lines showing the safe yield in million gallons per day of the water-sheds from which New York now receives its supply. The amounts represented by these lines have been carefully deduced as already described from the data obtained during the past 40 years from the Croton water-shed itself, and show the amount of water per day that can be secured in a very dry year. The lines representing the safe yield of the water-sheds should be observed very carefully. They are based on the actual records of the measured stream flow, making all allowances for evaporation, and for the effect of the reservoirs in holding back the surplus water, and represent the conditions on the Croton watershed with as great accuracy as it is possible to attain. They do not depend upon calculations from rain-fall, or on comparisons with other water-sheds, but refer precisely to the conditions that have obtained in the past on these very yielding grounds. The limit set for the Croton, Bronx and Byram combined, up to 1902, before the new Cornell dam was finished, was determined from the records of the low flows in 1879 to 1883, the driest period in the history of these works. The accuracy of these records was tested in 1901, when the draft by the City equalled the safe yield of the water-shed. That year had been one of small rain-fall in the spring months, with the result that the supply ran short and for a long period it was necessary to partially close the valves and reduce the pressure all over the City in order to pull through. Again, in 1901, a calamity was only fortunately averted by exceedingly heavy rains in the Summer and Fall. During 157 days, or nearly half that year, the City was obliged to draw from storage, the stream flow being insufficient to keep up the draft, with the R-sult that some 16,430,000,000 gallons was drawn out of a total capacity of some 44,000,000,000 gallons in all the reservoirs and ponds available in the Croton water-shed at that time. The total rain-fall during that year was about ten inches greater than the average for the forty-eight previous years, yet the uneven distribution, by which most of it fell during the Summer instead of in the Spring and Fall, was the cause of the depletion of the stream flow. An inch or two less of rain in the Spring of that year would have brought about a repetition of the conditions of 1891. 78 Waste of Water and Its Reduction. On the diagram referred to it will be seen that in 1901 the amount of water actually drawn from the Croton, Bronx and Byram water-sheds, was just equal to the amount it was safe to count on; and the close margin by which a calamity was averted that year, which ought to be fresh in the minds of New Yorkers, should be sufficient warning that these lines showing the amount which can be safely counted on for the supply of the City are placed on the diagram correctly. It will also be noted that at the present time the draft from the combined water sheds even with the new Cornell dam and the proposed Cross River dam in service exceeds the safe limit of permissible draft; and that if the City escapes a water famine in any year in the immediate future, unless vigorous and elaborate steps are taken immediately to stop leakage and wastage in the houses, it will be a blessing which we have now no right nor reason to expect. The condition is extremely unsafe and hazardous. On this diagram are also placed two dotted lines representing the probable amount by which the draft on the water-shed can be reduced by stopping leakage; one line showing the probable best effects obtainable by house to house inspections repeated twice a year with the infliction of penalties on those who neglect to keep their. plumbing in order, and the other showing what may be expected from the complete metering of the City in five years; a bigger task in this direction than has ever been attempted elsewhere. An appreciation of the conditions represented by this diagram will show the recklessness of attempts to defer the construction of the new water supply works now planned. Seven years is a short time to allow for the completion of such enormous works as these must be, and by seven years, with the most exacting and faithful work in suppressing leakage and wastage, including universal metering, the legitimate, actual needs of the City for water will not have been pulled down below the safe yield of the water-sheds, including the extra yield due to all the additional storage that can be made available by that time by the construction of the additional Cross River and Bronx reservoirs. Under no conditions would it be possible to secure continuous74 Part VII: Future Conditions. ly more than three hundred and sixty million gallons per day from the Croton water-shed; and to be able to secure this amount it would be necessary to provide, in adition to the storage reservoirs now in use, others with a total capacity one and one-third times greater than those we now have, including the new Croton Reservoir. By reference to the diagram again it will be seen that: First.-In about three years, unless wastage is stopped, the average daily draft by the City will equal the total amount of water it would be possible to take continuously from the entire Croton water-shed if every drop of collectible rain-fall wede stored in reservoirs two and one-third times as big as now exist. Second.-That for the next nine years, with the greatest amount of reduction in wastage and leakage that can be secured by the metering of every service in Manhattan and the Bronx, and using every possible effort to correct leakage in the street mains and service pipes, the legitimate demand for water in Manhattan and the Bronx will not be pulled down below the safe limits of the present supply even if increased by the building of the Cross River Reservoir and all the other emergency storage basins that can be constructed during that period. The conclusion to be drawn from proposition No. I is: Domestic consumers must be made to stop leakage and wastage from the fixtures on their premises as completely and as quickly as possible. The conclusion to be drawn from proposition No. 2 is: The work on the new water supply from the Catskill mountains must be started immediately and be pushed to completion as rapidly as possible. With the greatest practicable amount of leak reduction and waste reduction, the relief afforded will not justify the postponement of the building of these works one day. A third conclusion is: Although it will require a large outlay annually for the next eight or ten years to hold this wastage and leakage in check the expense can not be avoided and provision should be made fQr it at once so that a system of inspection can 75 Waste of Water and Its Reduction. be organized and put into immediate effect while waiting for the settlement of the question of the final control of wastage. The judgment of the present administration that an additional supply is necessary immediately and that a quantity not less than 250,000,000 gallons daily should be provided for at once, with the works so built that the aqueducts can take care of 500,000,000 gallons daily when necessary, is therefore sound and should be supported. While there may be differences of opinion as to whether this supply should be first secured from the Catskill Mountains, or from the water-sheds east of the Hudson River, or from the Hudson River itself, the general proposition that the increased supply is immediately necessary is beyond argument. The recommendations of the Burr, Hering, Freeman Commission for the development of the future Water Supply of New York, contemplate briefly the building of a large distributing reservoir on high ground near Mount Vernon, from which a new 500,000,000 gallon aqueduct is to be built northward to a point on Fishkill Creek, about 5 miles east of Fishkill, where it is to receive the waters of Fishkill Creek collected in a large storage reservoir. From this point a branch aqueduct is to be extended across the Hudson River to the Catskill Mountains to secure the 250,000,000 gallons per day that can be had from the proposed Ashokan Reservoir by building a large dam on Esopus Creek at Olive Bridge. The total amount of water provided for by this plan is 310,000,000 gallons daily, 250,000,000 from the Catskills and 60,000,000 from the Fishkill. When this supply will have become insufficient for the needs of the City it is proposed that Wappinger's Creek and Jansens Kill in Dutchess and Columbia Counties be developed by the building of storage reservoirs the waters from which are to be conveyed to the 500,000,000 gallon aqueduct ending at Fishkill Creek and thence to the City. The full development of these water-sheds east of the Hudson will provide about 260,000,000 gallons daily, which with the 250,000,000 from the Catskills would make up the 500,000,000 gallons sought as the amount required for such time in the future as it seemed necessary to make provision. 76 Part VII: Future Conditions. On Diagram No. 1 are placed lines showing the effect of the building of these different works on the total supply from all sources that will then be available. Table 17 (see next page) has been prepared from this diagram for conveniently comparing the conditions which will exist at different periods, and for the purpose of showing the length of time in the future provided for by the different supplies if wastage and leakage are not stopped as compared with those if universal metering is adopted. The Table is in two parts. In one part it is assumed that only Manhattan and the Bronx will be supplied from the new works; in the other it is assumed that the extra water needed in the future for Brooklyn, Queens and Richmond will also be supplied from the new works. Considering, first, only that Manhattan and the Bronx will be supplied from the new works it will be seen that the building of the Cross River reservoir which will add about 12,000,000 gallons daily to the supply will not help matters enough to keep the City out of danger as the demand for water at the present time exceeds the safe yield of the water-shed even if the Cross River reservoir were completed. It will also be seen that before the Fishkill supply can be made ready the demands for water by the City will exceed the safe yield of the Croton, Bronx and Fishkill combined, whereas, if the City is metered, to control leakage and wastage, the same works would take care of the consumption for about 13 years. Under this plan it is presumed that the Fishkill waters would be discharged into Croton Lake without waiting to build the large aqueduct to the City. If the Fishkill is not developed at present but the aqueduct is pushed at once to the Catskills the amount of water secured would last for about 16 years without waste reduction and about 30 years if the City is metered. After this, if the Fishkill were developed, the provisions would be sufficient for 20 years under present conditions and 36 years with universal metering. Then, with the final full development of all the water-sheds east of the Hudson the City will have been provided for until 1936, under present conditions, or 1956, with universal metering. In each case the 7T TABLE NO. 17. FUTURE PROVISIONS FOR INCREASED WATER SUPPLY OF NEW YORK CITY. ASSUMING NEW WORKS TO BB STARTED IMMEDIATELY. ASSUMING ALL NBW WORKS TO BB STARTED IMMEDIATELY. Brooklyn, Queens and Richmond not included, it being assumed that their additional supplies will be provided for independently. Brooklyn, Queens and Richmond included; assuming that all the extra water required for these boroughs in the future will be supplied from New York. SOURCE o0 SUPPLY. -s Go Number of years Number of years after complefrom date be- tion of new fore safe capaci- works before ty will be ex- safe capacity ceeded if will be exceeded if Amount of water that can safely be had per day, gallons. Cost of works in addition to value of present works. Number of years from date before safe capacity will be exceeded if Number of years after completion of new works before safe capacity will be exceeded if 40 a 1 -11 a s A t: 0. r. cl, o I. Waste is All servinot ces are stopped metered. Waste is not stopped. All services are metered. Waste is not stopped. All services are metered. Waste is not stopped. All services are metered. Croton and Bronx (present At the At the At the At the At the At the At the At the sources)........ present present present present 300,000,000present 0present present present time. time. timeime. tim. time. time. time. time... j. _ Croton and Bronx, witht Cross River dam added. At the At the present present time. time. At the present time. At the present time. 310,000,000 $1,500,000 At the present time. At the present time. At the present time. At the present time. _ Croton, Bronx, Cross River and Fishkill Creek...... Before completion. 13 Before completion. 8 370,000,000 19,500,000t Before completion. 12 10 Before completion. 5 3 -~,, I I Croton, Bronx, Cross River and Catskill supply....... 16 30 9 23 23 560,000,000 56,500,000* III I II I Croton, Bronx, Cross River, Catskill Mountain and Ti..s1t.ll& ro.r& 16 99 90 on or 1I O0 no0n nnn nnn A1 Ann nnn* 1 K Q -4 40 *o.o.. o vz*.*. *.**. U* z o U VVJt jVV V, 1,J VJI 1 VW, U V J J V VV LU. Croton, Bronx, Cross River, Catskill Mountain, Fishkill, Wappinger Creeks, and Jansens Kill. (This includes all the water available on East side of Hudson without pumping)............ 30 50 23 43 820,000,000 113,000,000* 24 41 17 34 t Based on discharging into Croton Lake. o * Based on aqueduct capacity for 500,000,000 gallons daily. Capacities and costs of Catskill, Fishkill, Wappinger and Jansen's Kill supplies based on Burr, Hering, Freeman Report, 1903; all other data from. Diagram No. 1. - Waste of Water and Its Reduction. number of years for which provision would be made as stated above date from the present year. If the extra water required for the future growth of Brooklyn, Queens and Richmond must also be taken from the same sources the practical effect would be to shorten the time for which provision would be made by from 4 to 9 years, depending on the period and on whether wastage and leakage are checked. A more detailed study of this Table will suggest several important conclusions that may be drawn, one of which, the financial aspect of metering, is discussed quite fully in Part IX to which attention is invited, it being therein shown that if all the Boroughs are provided for from the new supply, and if universal metering be adopted as soon as possible, the saving in cash annual maintenance charges, on the second installment of the increased supply, which will have to be provided and in use by 1918 if wastage and leakage are not checked, will, between 1918 and 1930, amount to a sum equal to one-half the cost of building these additional works when they become necessary, or two-thirds the cost if Manhattan and the Bronx only are supplied from the new works. 80 PART VIII. PRACTICABLE METHODS OF CONTROLLING LEAKAGE AND WASTAGE IN NEW YORK. IN order to carry the City safely through the years that must elapse before an additional supply of water can be had it is absolutely necessary to stop the leakage now prevalent all over the City in unmetered buildings. As the City has not at the present time the right to meter domestic consumers inspections and penalties must be depended upon until the question of final control of leakage and wastage can be settled. It will certainly take a year to secure the necessary authority and no definite results can be expected from metering until a year after the installation of meters has commenced. It is therefore imperative that immediate action be taken. The Commissioner of Water Supply, Gas and Electricity, and the Chief Engineer, Mr. I. M. De Varona, are awake to the gravity of the situation and have already asked for appropriations and authority to put a system of inspection in force. The plan proposed is substantially that which has been outlined, viz., the division of the City into districts, the examination of the flow in the street mains to discover leaks and locate properties where leakage and wastage are prevalent, the hunt for illicit water-takings and the infliction of penalties on premises where leakage or wastage are discovered on second inspection. This movement should have the earnest and active support of every householder, owner and tenant alike, as in no other way can immediate relief be had from the probability of a water famine. This method of controlling leakage and wastage, however, must be understood to be only a temporary expedient the results of which will be lost as soon as discontinued and the effectiveness of which will not compare with that to be derived from the placing of meters on the services of all consumers. The total cost of the inspection system, for which the consumers must ultimately pay, will almost reach the cost of maintaining a meter system; and with meters once installed there is no danger of the future dis81 Waste of Water and Its Reduction. continuance of their use, while the results obtained will be permanent and the question of the suppression of controllable leakage and wastage will have been settled for all time. But until authority can be obtained to adopt universal metering the inspections will have to be depended upon regardless of cost or permanent benefits. It is difficult to estimate exactly the cost of such a system of inspection as much depends upon the thoroughness with which it is done, and on the methods employed. In some sections it will be more expensive than in others; and the City is so large and the conditions are so complex that the cost of doing the work can only be found out by trying it on a large scale. As a start, I estimate that for effective work the cost for the first year would be in the neighborhood of $200,000. This sum would provide a fund of $50,000 for purchasing the necessary apparatus and supplies for testing the mains and the making of permanent connections through which the mains could be tested in after years without the necessity of digging up the streets, The balance, $150,000, would provide a staff of house-inspectors, main inspectors with assistants, laborers, a general superintendent, general foremen, book-keepers, clerks, stenographers, as well as sufficient reserve to cover the cost of postage, stationery, hauling, and car-fares. For this outlay I believe that the City could be efficiently covered once a year for street main inspections, and twice a year for the house inspections. In subsequent years the cost could probably be reduced to about $100,000 per year. This is at a considerably smaller rate than the actual cost of the tests made by Mr. Hill, but it is believed that with careful supervision, after the first year's test of the mains to locate the most serious leakage and most extravagant wastage the street main work could be very much reduced, greatest dependence being placed on the house-to-house inspections to keep plumbing in order; and the greater experience of the men and smaller proportion of probable leakage will tend toward greater efficiency and reduced expense. If the street main work is omitted entirely an appropriation of $40,000 to $50,000 per year should be sufficient to cover the cost of 'the house-to-house inspections. There is little doubt, however, that the discovery of underground leakage, both in the 82 Part VIII: Practicable Methods of Control. mains and service pipes, and the discovery of large unknown connections, which can only be detected with the aid of the street main tests, will amply repay the extra cost of this important work. If it could be assured that the necessary appropriations would be forthcoming each year for the systematic carrying on of this system of inspection there is no doubt that a sufficient amount of wastage could be controlled to carry the City through until the completion of the new works without much likelihood of more serious drawbacks than possibly the throttling of the supply during one or two seasons if the stream flow should be much below the normal. But, owing to changes of administration, which must occur, and to the relief from the scare if no actual shortage occurs in a previous year, it would only be history repeating itself if the work were stopped; and then, in a year or two, matters would again be in as bad a shape as before. On Diagram No. 1 it will be noted that the dotted line representing what I should consider the possible reduction of the consumption as a result of house-to-house inspections keeps, after dropping to the lowest point, about parallel with the curve of total necessary use up to the time the new works will be ready (if construction is started as soon as possible) and then again rises to the original rate of consumption prevailing before waste reduction was undertaken. There is no question but that this would be the result: as soon as the additional supply becomes available all efforts to prevent waste will cease, if dependent upon inspections, and the consumption will jump up again to figures that would have been reached if no effort had been made to check it. This has happened before in New York; it has happened in Newark, Boston, Detroit, and, in fact, in every city that has grappled with this question. Under the inspection system the nine years' work will have represented a total cost in Manhattan and the Bronx of about $2,000,000 and the expenditure will have been merely on an expedient to pull through a hard time which when passed will leave no permanent benefits behind to show for the work done and the money expended. So much opposition is generally raised when the meter question is agitated that it may be well to answer in advance the arguments most frequently used against their adoption. 83 Waste of Water and Its Reduction. FIRST OBJECTION. The undesirability of restricting the use of water. ATER should be supplied in the most lavish abundance and instead of restricting its use, every inducement should be held out to encourage a greater, or more lavish use than is now prevalent; and for this purpose the meter is the most effective agent we have to-day. The meter is not used to restrict use but to find waste and leakage the stoppage of which will make available for use the large amount of water running to waste to the sewers through leaky fixtures. It is never proposed to install meters for the purpose of discouraging the use of water; and, when rates are properly adjusted metering never has this effect. In Atlanta, Ga.; Richmond, Va.; Detroit, Mich.; Newark; N. J.; Asbury Park and Atlantic City, N. J., metering was undertaken for the express purpose of making available for use the water which had been previously allowed to leak away. The results were: Atlanta, Ga.-The limit of the supply had been reached in 1884. Meters were rapidly installed on all services and it was nine years later before the total amount required for the supply of the city was as great as it had been in 1884. No restrictions were placed on the use of water, but leakage was stopped; the amount of water used per capita is still high, about twice the amount used in Fall River, Mass. Richmond, Va.-The consumption of water per capita daily in Richmond is about as great to-day as that in New York City notwithstanding the fact that nearly all the manufacturing and about three-fourths of the domestic consumers in Richmond are metered. The use of meters has not very materially reduced the amount of water consumed per capita, but it has resulted in increased pressure all over the city, giving an ample supply to all those in the high level districts who previously frequently could get no water. In other words, use has not been curtailed, but leakage has been stopped, at least from the majority of fixtures in the residences. There is still considerable waste in the water used for public purposes. 84 Part VIII: Practicable Methods of Control. Atlantic City.-In 1896 the average daily consumption was 5,259,000 gallons. Meters were placed on about three-fourths of the services during the next two years. It was not until after the lapse of nine years that the average daily consumption reached the amount used in i896, the resident population having increased 50 per cent and the summer population not having changed materially during this period. The average daily use per capita for the three summer months has gradually increased from 35 to 58 gallons, thus proving no restriction of use but control of leakage and waste. Asbury Park, N. J.-The experience at Asbury Park has been similar to that at Atlantic City except that being a much smaller place and growing at a rapid rate the period of relief from water shortages was about 4 instead of 9 years. Detroit, Mich.-A house-to-house inspection coupled with the placing of meters on premises where excessive leakage and wastage were discovered resulted in stopping inside leakages in such great amounts that this city, which had in the meantime increased in population from 188,000 to 269,000, after the expiration of the 9 years used no more water for its total supply than before the stoppage of leakage was begun. During this entire period not over about 10 per cent of the services in the city were metered, hence the reductions were accomplished mainly by inspections and therefore the increase of available supply consisted solely of the water saved by the prevention of leakage inside the buildings. This is the precise result that is obtained by metering. Newark, N. J.-The results in Newark were along the same lines as those accomplished in Detroit, but the campaign was only continued for two years and the proportionate reduction of leakage and wastage less. SECOND OBJECTION. Alleged injurious effect of meters on public health. T HIS objection is generally made the most of by owners of property who fear that if meters are installed their properties will be saddled with liens for the water wasted by tenants, and predictions of typhoid fever epidemics and all sorts of d'ias85 Waste of Water and Its Reduction. ters are used as arguments to compel tenants to support the owners in the campaign against waste reduction. The death rate from typhoid fever is the surest indication that can be had of the purity of a water supply so far as its ability to transmit water-borne diseases is concerned. Hence, if meters have any connection with the prevalence of typhoid fever their effect should show in the death rate of the cities using them extensively. Viewed in this light the statistics given in Tables 18 and 19 are pertinent. TABLE NO. 18. TABLE SHOWING TYPHOID FEVER DEATH RATE PER 100,000 FOR SEVERAL YEARS IN CITIES WHERE A LARGE PERCENTAGE OF THE SUPPLY IS SOLD BY METER. YEAR. CITY. 1890 1891 1892 1893 1894 1895 Atlanta..... Typhoid death rate....... 149 119 87 66 43...% of services metered.....00 100 100 1 00 100 Fall River. Typhoid death rate.......62.. 38 20 29 33.. %of services metered..... 75 76 77 78 80 82 Lawrence... Typhoid death rate...... 123 115 95 69 48 31... % of services metered..... 29 34 39 45 50 56 Lowell...... Typhoid death rate....... 82 96 90 62 60 39..... % of services metered..... 22 24 26 28 30 33 Milwaukee.. Typhoid death rate....... 33 33 31 37 26 25.. % of services metered..... 33 40 43 48 51 54 Providence.. Typhoid death rate....... 29 47 36 34 49 31 % of services metered..... 62 63 65 69 72 75 Berlin...... Typhoid death rate....... 9 10 8 9 4 "...... % of services metered.. 87 88 88 86 86 Breslau.. Typhoid death rate....... 15 12 15 10 6.... % of services metered..... 77 77 74 75 75 Munich..... Typhoid death rate....... 8 7 3 15 2.5.... of services metered... 90 92 93 94 95. 86 Part VIII: Practicable Methods of Control. TABLE NO. 19. TABLE SHOWING TYPHOID FEVER DEATH RATE PER 100,000 FOR SEVERAL YEARS IN CITIES WHERE BUT A SMALL PERCENTAGE OF THE SUPPLY IS SOLD BY METER. YEAR. CITY. 1890 1891 1892 1893 1894 1895 Brooklyn.... Typhoid death rate...... 21 21 18 18 16 16 Chicago...... " Is.... 83 160 103 42 31 32 New York.... i 44.... 21 22 14 21 17 17 Philadelphia..... 64 64 34 41 32 St. Louis..... 34 30 37 103 31 19 Edinburgh......... 19 18 13 14 15 Glasgow..... 26 31 18 20 24 Hamburg.......... 28 23 34 18 6 London. 16 15 11 16 15 From the above tabulations it will be seen that water meters have no effect, either for good or for bad, on typhoid fever death rates in cities. Some unmetered cities have death rates as high or higher than metered cities. In Lawrence, with the increase of meters the death rate steadily dropped; in Atlanta, with all services metered throughout the whole period the rate also steadily dropped. The rate in Brooklyn with practically no meters is less than in Atlanta where all services are metered; the rate in Fall River, practically all metered, is less than in Chicago and Philadelphia, which are practically without meters. In other words, meters have no effect whatever one way or the other on the typhoid fever death rate of a city. The same is true regarding the death rates from all other diseases and regarding the total deaths from all causes. There is no traceable connection whatever between metering and public health to be found in the records of any city in the world. It is sometimes stated that the use of meters will restrict the proper and lavish consumption of water and that this will have a tendency to lower the general health tone of the people even if it does not actually cause deaths. Such a statement is all theory with no facts to back it up. Assuming, for the sake of argument, that the use of meters 87 Waste of Water and Its Reduction. would actually restrict the use of water, that many people would have to be satisfied with much less water than they now use, a condition which distinctly would not result if meter rates were properly adjusted, there is no support for the theory advanced that the general health tone of the community would be lowered. In Berlin the total amount of water supplied for all purposes is about 18 gallons per capita per day, while in New York the average daily supply is about 135 gallons, or over 7 times as much; yet the general health tone of Berlin is better than that of New York in practically every particular. The typhoid fever mortality in New York averages from three to four times as high as in Berlin and the total death rate from all causes about 20 per cent. higher. Berlin is completely metered and has a very small per capita supply of water, while in New York scarcely any domestic consumers are metered and the per capita supply of water is about 7 times as great. Similar comparisons could be made for other cities. THIRD OBJECTION. The allegation that the installation of meters will lead to reduced receipts from the sale of water thus causing a deficit in the revenues'and hence making an increased tax levy necessary. HIS is purely a matter of business administration. A certain amount of money must be raised each year to pay the operating expenses of the Water Department, regardless of whether the water is sold by frontage rates or meter measurement, and can be taken care of by proper adjustment of meter rates. Where these rates have been properly fixed no deficit is possible and therefore no increased taxation will be necessary. The only city of which I have knowledge where it has been necessary to make appropriations to cover the loss of revenue resulting from metering is Milwaukee. During the last two or three years this has been necessary in that city owing to the fact that a uniform rate of 6 cents per thousand gallons was made for all metered water, manufacturing and domestic, regardless of quantity, with no minimum rate except a charge of $1.00 per year for 88 Part VIII: Practicable Methods of Control. reading the meter. Under this rate a surprisingly large number of consumers paid less than $3.00 per year for their water, and a great many less than $2.00, and the revenues fell below the expenses. Even this, however, worked no hardship as the small appropriation to cover the deficit was no more than would have come out of the consumers if paying on frontage rates, or meter rates adjusted to produce the necessary revenue. In cities where the entire supply is pumped the saving in cost of fuel resulting from stopping leakage and wastage frequently will amount in one year to more than the cost of the meters and their installation. In New York (Manhattan and the Bronx) although a considerable amount of the supply must be pumped to the high level districts, this has not been considered in the calculation showing the benefits from metering as given in Part IX. In the matter of the adjustment of rates it would seem, should it be desirable and possible to put the City on a meter basis, that a combination of frontage and meter rates would be the most equitable system. A certain amount of water must be supplied to the City in excess of that for which the meters will account and this water and that used for fires, street sprinkling, sewer flushing, etc., form together a portion of the supply from which no direct revenue can be collected in an equitable manner by meter rates. Yet it is impossible to dispense with this water, and as all property as related to that portion used for public purposes is benefited directly and, as related to fire insurance rates, indirectly, it would seem more equitable to provide by frontage rates revenue sufficient to cover its cost and charge by meter rates only for water consumed by actual use. It would also seem advisable in New York, in view of the large amount of water used for manufacturing, business and commercial purposes, to have a sliding scale of charges for metered consumers, giving the large consumers a lower rate than the small consumers and establishing a minimum rate to make it possible to maintain the system in good condition financially. For this minimum rate the consumer should be entitled to the use of a certain quantity of water, the amount of which, as well as the rates of charges, to be determined after investigation. For amounts used in excess of the stated quantity meter rates should be paid. 89 Waste of Water and Its Reduction. FOURTH OBJECTION. The fear that the adoption of meters would put the city in the grip of the meter trust. T HIS objection to meters is a new one, originating in a resolution adopted March 15th, 1906, by the Philadelphia City Councils. There are now in use, in New York City, about 40,000 meters of well-known standard types, each of which has been tried and tested by actual use for many years and has been found satisfactory by the Department of Water Supply. Under competitive bidding the City would receive independent proposals from the makers of practically all the types of meters now in use. The Meter Trust, if there is one, would have pretty lively competition to get the contracts and the City would get the benefit. In any arrangement under which meters would be used in New York City, the City should purchase, own, set, inspect, repair, test and reset every meter, and should provide for purchasing at a fair price all meters now in use. In no other way can proper supervision be had over the meters and accounts. In no other way will it be just to the consumers if universal metering is adopted. The meters should be bought under competitive bidding governed by rigid specifications carefully prepared by the City, in the same manner as other machinery or supplies, requiring the meters to fulfil certain prescribed tests as to accuracy, durability, reliability, back registering, choking, hot water tests, etc., according to the requirements for the different purposes for which they will be used. The City should have a modern and well-equipped repair shop where a stock of meters and parts can always be kept on hand and where meters may be taken periodically for testing. The meter rates should then be adjusted at such figures that in paying for the water used the cost of the meter, and the cost of setting it, including the cost of repairs, testing and renewals, would be distributed over a period corresponding to the life of the meters, say some 20 years. 9o PART IX. THE INTRODUCTION OF METERS IN NEW YORK AS A BUSINESS PROPOSITION. T the present time practically all the large consumers in Manhattan and the Bronx have been metered, so that in extending the system to all water takers very few large meters will be required for several years. It is, of course, impossible to forecast how many meters of each size will be required in any year, and as the prices of meters vary from about $6.50 for a /s-inch, to about $400 for a 6-inch meter, there is a possibility of wide divergence between two independent estimates. In Manhattan and the Bronx owing to the absence of curb stop cocks, and to the quick removal of snow during the Winter, the placing of all meters outside the buildings in special settings would be so expensive as to be almost prohibitive, necessitating the digging up of all the services, placing cocks therein and build; ing double-covered, frost-proof vaults or settings. I have therefore assumed that all meters will be set in basements, the larger sizes in wooden boxes for protection. As a basis for this estimate I have assumed that of the required total number, 75 per cent would be Y/8-inch meters 15 " " " 34 6 '" " " 1 2, " " " 152 1 " " " 2 " 1 " " ' 3 and larger and have estimated the cost of these sizes and the cost of setting as shown in Table 20. 91 Waste of Water and Its Reduction. TABLE NO. 20. COST OF METERS AND SETTING. Cost of Setting, SIZE. Cost of Including Materials Total Cost. Meters. and Labor. I inch..................... $6.50 $4.00 $10.50 i "..................... 16.00 6.00 22.00 1 "'.................. 27.00 8.00 35.00 11.................... 37.00 10.00 47.00 2..................... 45.00 12.00 57.00 Average 3-inch and larger.. 90.00 13.00 103.00 On the basis of the percentages of the different sizes estimated to be required the average cost, at the above prices, would be $15 per meter in place. This figure is probably as close as can be arrived at without a greater amount of labor than the available time will allow. Probably, in view of the great number of meters required and the large, well-organized force necessary to place them in a short enough time to quickly derive full benefit from their use the prices adopted may be somewhat higher than necessary. A measure of reliability is given to the average cost thus derived by comparison with the actual average costs of the meters in use in other cities, of which the following are a few examples. TABLE NO. 21. AVERAGE COST OF METERS IN USE. Average Cost of all CITY. Meters in Use, In- Remarks. eluding all Sizes. Atlanta, Ga... $12.15 Atlantic City, N. J. 15.19 Cleveland, O......... 18.25 Mostly outside settings, many in brick vaults. Detroit, Mich........ 29.00 Practically all on non-domestic consumers. Harrisburg, Pa....... 13.50 Hartford, Conn...... 19.20 All {-inch meters and larger. Madison, Wis........ 15.71 Providence, R.I...... 16.50 The annual expense of metering would be made up of the interest on the cost of the meters, a depreciation charge sufficient to 92 Part IX: Meters as a Business Proposition. pay for the meter in about 20 years (its assumed life), the cost of inspection, repairs, renewals of parts and testing, and the cost of reading the meters, delivering bills, etc. The cost of repairs and renewals varies from 1 per cent. to 6 per cent. of the value of the meters in use, according to the local conditions in different cities. TABLE NO. 22. ANNUAL COST OF REPAIRS, ETC. Annual Cost of Repairs, Renewals, Testing, and MainCITY. tenance for Several Years; Per Cent. of Cost of Meters in Use. Atlanta........... About 4 to 5 per cent. Cleveland......... " 23 Fall River......... " 1 Harrisburg......... " 1 Hartford.......... " 1 to 6 " average about 3 per cent. Providence........ " 2 " I have assumed for New York conditions, owing to the probability that some of the meters in exposed places may freeze and break during very cold weather, an allowance of 33 per cent. of the value of the meters in use to cover this item. The cost of reading the meters, sending out bills, etc., will depend on the organization of the force and the frequency with which the meters are read. Assuming quarterly readings an inspector should be able to take care of about 1,500 meters at a cost for labor, clerical work and delivering bills of about $1,000 per year, or say 66 cents per meter per year. This would make the total annual cost per meter per year as follows: Interest on cost of meter, $15.00 at 3%o......... $.45 Depreciation, life 20 years.....................75 Maintenance, renewals, testing, repairs, etc., $15.00 at 33..........................50 Reading meters, sending out bills, etc...........66 Total cost per year per meter............ $2.36 93 Waste of Water and Its Reduction. This figure includes the cost of the meters spread over 20 years and the yearly interest on the cost of the meter in place, so that if provided for out of a bond issue the cost given per meter will take care of the retirement of the bonds at maturity. The direct value of meters for New York conditions can be found by calculating the saving resulting from their use in making it possible to defer the date of constructing the second installment of the additional works after the first installment of the proposed Catskill supply shall have been completed. If Brooklyn, Queens and Richmond, as well as Manhattan and the Bronx, are to be supplied from the new works, and if no method is put in force for the control of wastage, the limit of the capacity of the first installment of the new proposed Catskill works will have been reached by 1918, whereas if the City is put on the meter basis the additional supply will not be needed until 12 years later. The cost of the second installment of the additional works will. be $51,400,000.00 if the water-sheds on the east side of the Hudson are developed to 260,000,000 gallons daily capacity, to include all the water obtainable without pumping from Fishkill Creek, Wappinger's Creek and Jansens Kill. During these 12 years, with no meters in service, the annual cost will have been as follows: 904 Part IX: Meters as a Business Proposition. Interest on $51,400,ooo.oo at 3 per cent..................... $1,542,000.00 Sinking fund to retire bonds in 40 years,-$51,400,000.00 at 1.37 per cent........................................... 700,000.00 Depreciation on gates, valves, pipes, filters, etc., at 6 per cent.. 310,000.00 Ordinary repairs, keepers' wages, sanitary policing, laboratory, and accounting.................................. 90,000.00 Cost of filtering wasted water, I50,000,000 gallons daily at $3 165,000.00 Total annual expense...................................... 2,807,000.00 With this should be compared the cost of metering during the same period. The average number of services in use for the period, at a somewhat less favorable ratio than obtains to-day, would be: Average population during period, 4,800,000 Average number of persons per service, 16, = 300,000 With a meter on every service the cost per year would be: 300,000 meters at $2.36 per meter per year for meters, repairs, renewals, interest, reading, sending out bills, etc. 708,000.00 Difference, representing excess cost per year, due to waste and leakage......................................... $ 2,099,000.00 or, for the I2 years................................... $25,188,000.00 (ABOUT HALF ENOUGH MONEY TO BUILD THE ADDITIONAL WORKS WHEN THEY BECOME NECESSARY) If Brooklyn, Queens and Richmond are supplied from other sources and only Manhattan and the Bronx from the proposed new Catskill works the comparison is still more favorable, as the construction of the second installment of the works could be deferred about 15 years, resulting in a cash saving to the City after paying for the meters, their repairs, testing, renewals, maintenance, reading and sending out bills, etc., of about $32,000,000, or almost two-thirds of the amount necessary to build the second installment of the new works for a yield of 260,000,000 gallons daily. 95 Waste of Water and Its Reduction. To state this a little more in detail: If waste is not gotten under control the City will have paid out, in 15 years, in annual charges on an increased supply, after the limit of the first installment of the Catskill supply has been reached, a total of.................... $42,105,000.00 While, during the same time, with the use of meters to postpone the construction of these works, the total amount paid out will have been............................... 10,600,000.00 Showing saving by the use of meters in the 15 years of..... $31,505,000.00 Which will pay two-thirds of the cost of the next large necessary addition to the Water Supply of New York, if secured, as it logically should be, by developing the water-shed on the east side of the Hudson tributary to the new aqueduct. The use of meters, therefore, instead of bringing about an increase of taxation will save the City a sum of money that will go far toward providing the next increase in the supply, or if handled in another way, would permit a reduction in the water rates that would have to be collected, if wastage were allowed to continue, equal to about $2,000,000 per year for that 15-year period. New York City has few opportunities of making investments that will show such a large return. The cost of these meters, their maintenance and repairs, renewals of those broken and worn out, inspections, testing and sending out the bills, makes a total of $10,000,000 for which the City receives a clear profit of $30,000,000 in 15 years, or 300 per cent., or 20 per cent. per year. While no man living can with certainty predict the precise conditions that will obtain twenty years in the future, or ten, or even five years, either as to the growth of the population, or the rate of water consumption, or the possible extent of waste reduction, it is unnecessary for present purposes that these predictions should be exact; for it matters not whether the postponement of the construction of the second installment of the new supply be for two years or twenty, or whether the meters be completely paid for as fast as introduced or be gradually paid for out of the proceeds of a bond issue running 20 or more years, the adoption 96 Part IX: Meters as a Business Proposition. of universal metering will be a paying policy in New York City whenever undertaken in the future. This is shown in Table 23, which exhibits the probable saving to the City, as the result of universal metering, corresponding to the postponement of the building of the second installment of the proposed new supply for different assumed periods varying from 1 to 15 years. Up to the present time universal metering has not been urgently necessary in New York, and it may even be questioned whether from a practical point of view it would hitherto have been advisable; but the situation is now entirely changed. Subsequent new additions to the supply must provide for larger and larger quantities, each more difficult and more expensive to acquire than its predecessor; and the day is not very far distant when the only available remaining source will be the Hudson River, the waters of which will then have to be secured by pumping from some point beyond Poughkeepsie or be brought by gravity in enormous aqueducts from the Adirondack Mountains 200 miles away. When a new gravity supply is introduced it costs just as much per year for interest, sinking fund and operation expenses whether all, or only one-tenth, of the amount of water thus made available is consumed. Therefore property owners must always pay for more water than is actually used, the disproportion between quantity and cost being greater in the earlier years, and being in direct proportion to the magnitude of the excess provided. Where leakage and wastage are not under control this excess necessarily has to be made materially greater to provide for the future growth of the City than where the supply is properly husbanded, otherwise the extensions of the works would necessarily become much more frequent; and either alternative would greatly increase the taxes that would have to be levied to provide for the maintenance of the works. In the near future, owing to the dangers due to the pollution of the streams, dangers which can not be entirely forefended by any practical protective measures, the water supply of New York will require filtering to safeguard Public Health. Filtering costs money; and the wasting of water which has been prepared by this process is an entirely different proposition from wasting the 97 to TABLE NO. 23. Showing that the Postponement of the Construction of the Second Installment of the New Water Supply for even two years, as the Result of Universal Metering, would be a Paying Policy even though all the Meters were fully paid for during those two years; and that the longer the Period of Postponement the greater the Saving. If meters are not used, and The total The aggre- wastage prevails, Saving, resulting If the building of cost of repairs, gate cost of the second install- from universal meter- Saving, resulting the Second Install- The aver- The t o t a 1 renewals, in- meters, fully ment of supply ing, assuming all meters from universal meterment of the New age number c os t of the terest, reading paid for dur- will have to be in to be fully paid for, dur- ing if cost of meters be Supply, which, un- of Meters in meters in use, meters, etc., ing period stat- use, and the main- ing period stated in Col- provided for out of bond less wastage is under use during at $15.00 each, for entire peri- ed in Column tenance charges umn 1, as well as all re- issue running 20 years! control, willhave to the period fo r entire od stated in 1, and for me- thereon, compris- pairs and maintenance instead of paying for be ready by about stated in period stated Column 1, ter mainte- ing interest, depre- charges due to the meter them completely durin 1922, can, by uni-Column 1 in C o lum n would be at nance, during ciation, sinking system's adoption, periods stated in Colversal metering, be would be 1 would be $1.61 per met- same period, fund and operat-would be, for entire umn I. er, per year would be ing expenses, period stated in Column - would be, for peri- 1, od stated in Column 1, ________ Postponed 1 year 219,000 $ 3,280,000 $ 362,000 $3,642,000 $ 2,800,000 Loss $ 842,000 Saving $ 2,282,000 t 2 years 222,000 3,330,000 714,000 4,044,000 5,600,000 Saving 1,556,000 4,552,000 3 225,000 3,375,000 1,086,000 4,461,000 8,400,000 3,939,000 6,804,000 ~ 4 " 227,000 3,405,000 1,460,000 4,865,000 11,200,000 6,335,000 9,052,000 5 230,000 3,450,000 1,850,000 5,300,000 14,000,000 8,700,000 11,280,000 0. 6 232,000 3,480,000 2,240,000 5,720,000 16,800,000 11,080,000 13,512,000 7 234,000 3,510,000 2,650,000 6,160,000 19,600,000 13,440,000 15,729,000 * 8 " 237,000 3,555,000 3,050,000 6,600,000 22,400,000 15,800,000 17,920,000 9 " 241,000 3,615,000 3,470,000 7,085,000 25,200,000 18,115,000 20,070,000 10 " 244,000 3,660,000 3,930,000 7,590,000 28,000,000 20,410,000 " 22,220,000 11 247,000 3,700,000 4,370,000 8,070,000 30,800,000 22,730,000 24,376,000 12 250,000 3,750,000 4,830,000 8,580,000 33,600,000 25,020,000 26,520,000 13 253,000 3,810,000 5,300,000 9,110,000 36,400,000 " 27,290,000 28,630,000 14 256,000 3,840,000 5,770,000 9,610,000 39,200,000 " 29,590,000 30,750,000 " 15 " 259,000 3,880,000 6,260,000 10,140,000 42,000,000 " 31,860,000 " 32,830,000 Part IX: Meters as a Business Proposition. surplus water from a gravity supply which costs just as much per year whether part of it is used and part wasted or whether all of it is used. There is, for American conditions, no way to stop this loss except by universal metering. When considering future probabilities universal metering for New York City thus assumes great economic importance, not only for its immediate financial benefits but also for its material effect on the future growth and importance of the City. Unless leakage and wastage are brought under control there will eventually be locked up in the New York's water supply works hundreds of millions of dollars of the Public's money which under proper conditions should be available for business and commercial uses. The postponement of the construction of the second installment of the proposed new supply for 15 years would alone represent an average saving of about $2,000,000 per year, which invested in business with a return of only 3 per cent per annum would represent in that 15 years a total value of about $38,500,000, of which $30,000,000 would represent the total principal saved and $8,500,000, the sum of the progressive earnings year by year of the increments of the principal. New York City, while the most important, is only one of many cities which must for all time depend for water upon the streams tributary to and forming the Hudson River. All the cities on this water-shed are growing in population; and the increasing difficulties of supplying water to them as one watershed after another is exhausted, and the hardships imposed upon certain municipalities, by others, through the appropriation of waters nearby which may ultimately be needed in the future for their own water supplies or for the development of manufacturing industries, make it imperative that communities taking water from without their corporate limits should, as a matter of public policy, use this water economically and properly instead of allowing a significant percentage of it to leak away into the sewers. It was stated in the first pages of Part 1 that the problem of waste and leak reduction is one of convincing owners of property in New York City that it is to their interest to assist in stopping these losses, which add materially to their taxes year by year, by showing: 99 Waste of Water and Its Reduction. First.-That leakage can be stopped quickly and that when once stopped, can be permanently controlled. Convincing proof that leakage and wastage can be stopped has been offered in Part VI. From the data contained therein it will be seen that either house to house inspections, partial metering, combined with inspections, or complete metering will be effective for the purpose, but that permanent control of wastage is not obtainable for American conditions except by universal metering. Second.-That owner of properties run no risk of being robbed by wasteful tenants when the water is sold by meter. The answer to this is found in every city where meters are in use. Berlin affords one of the most instructive examples, as in Berlin a larger percentage of consumers are tenants than in New York, and the meter system has been voluntarily extended by the landlords; it saved them money, therefore they adopted it. Where meters are largely in use and where the meter rates are properly adjusted, our American cities all tell the same story. Those few in which domestic consumers are metered only on request, and in which the system has not rapidly extended, can find the reason therefor not in the wastefulness of tenants, but in the stupidity of the authorities in sticking to inequitably adjusted rates. Where rates are properly adjusted every owner will demand a meter when he sees his neighbor's water-bills reduced. The best evidence of the fact that tenants do not rob landlords, however, is found in the records of our metered cities, in all of which the majority of consumers are tenants, and in practically all of which the water rents are liens on the premises served, and yet there is not one of these that would go back to assessment or frontage rates. When rates are properly adjusted the sale of water by measure is the fairest, most satisfactory and most economical method, not alone for the owners, but for the entire city. Third.-That it is to the financial advantage of the owner to keep the plumbing in his buildings properly repaired and pay for the water used rather than to neglect to keep his property up and pay for the water wasted. 100 Part IX: Meters as a Business Proposition. This proposition is answered completely in Part IX, wherein it has been shown that with leakage and wastage under control the postponement of the construction of the second installment of the Catskill supply will save about $2,000,000 per year above the cost of metering for from 12 to 15 years. During that period there will be some. 300,000 properties in New York City which will require attention. It is not necessary in this connection to take account of accidents such as the bursting of service pipes by freezing, etc., as these would have to be repaired in any event, nor of general depreciation, which will go on and must be taken care of whether meters are used or not; the only repairs necessary to be considered are those due to worn-out fixtures and leaky valves, principally in water closets, which for an average expense of $1.50 to $2.00 per year per service could be kept from leaking after once having been put in proper order. This seems like a high figure when it is realized that there are thousands of buildings in New York where for years no leakage has taken place and not a cent has been spent for repairs. Perhaps a more rational way to look at it would be to consider that the 15 per cent. of the properties from which most of the leakage takes place would have to be overhauled every year (a different 15 per cent. each year) and be provided, at an average cost of, say, $10.00 per year per property, with new faucets and ball cocks to replace those worn out and leaking. This would represent a total cost of about $450,000 per year to be spent by owners for stopping leaks in the entire City. Under the assumption that the plumbing averaged of about the same durability all over the City, each owner would then have to spend $10.00 in 6 years to stop leaks, or about $1.66 per year. Even if the average cost were 7 times this much metering would still pay, for without metering the water taxes during the period referred to will have to be about $2,000,000 per year higher than if universal metering is adopted in order to cover the annual charges for the second installment of the Catskill supply, which will then have to be in use if leakage and wastage are not stopped. Therefore there will be no justification whatever for threats by landlords that if the meter system is adopted the rents will have to be raised. The adoption of universal metering will, with properly adjusted rates, save landlords money and afford no excuse for increasing rents. Everybody can have as much water 101 Waste of Water and Its Reduction. as is now used for no more, and probably in 85 per cent. of the cases for a little less money than is now being paid annually. There will be no necessity for being economical in the use of water under the meter system; it will be used just as extensively and with the same lavishness as to-day, but there will be no leakage and those who wish to waste it can do so by paying for what they waste, and those who do not will not have to help to pay for that wasted by others. Another way by which wastage can be controlled with the aid of meters is to put a meter on every service at the City's expense, maintain a force of meter readers and inspectors, the same as if the City were permanently on the meter basis, but let the water rates remain as at present and penalize with nominal fines properties where the meters show waste or leakage. This would have pretty nearly as great a restraining effect on waste and leak control, but the system would cost quite a little more to maintain than a straight meter system and would not afford that equitable distribution of charges, whereby each consumer pays for what he uses, that forms one of the chief attractions of the plan under which all water is sold by measure. The net advantage of this modified meter system would be less than of the simpler one of universal metering, although it would without doubt enable waste and leakage to be kept largely under control; it was used successfully for a number of years in one of the suburbs of London. Its disadvantage is that the consumer gets no benefit from being careful, but gets fined if he is careless; that is, it is all disadvantage and no advantage from the consumer's point of view, and has all the other disadvantages of the house-to-house inspection method, including the probability of its discontinuance by subsequent Administrations. It is plain, therefore, that there is no half-way ground in this question. It is purely a matter of which policy the City wishes to follow in the future. If the property owners wish to perpetuate the past policy of beginning work on a new water supply plant as soon as the one now proposed is completed, and so on to eternity, their taxes going up each time in consequence, then the method of house-tohouse inspections will tide over the present emergencey, but very 102 Part IX: Meters as a Business Proposition. probably at the expense of reduced pressures during some of the years. If they wish to get the full benefit of a good business proposition, whereby their taxes can be kept at the lowest possible limit, then universal metering should be adopted as soon as the necessary right can be acquired to place meters on domestic consumers. In conclusion I desire to state that owing to the limited time available for preparing this report it has been impossible to quote authorities and references in detail for all the data used, but I wish to express my great appreciation of the kindness of the many superintendents and managers of water works plants who have furnished data used herein; and I have included the titles of the reports from these various cities, with the names of the officials furnishing the data, in Appendix E, which contains references to the works and special articles consulted in connection with this investigation. The data regarding the stream-flows of the Croton and Bronx water-sheds since 1899 have been taken from the Annual Reports of the Department of Water Supply; those for prior years from the recalculations of John R. Freeman, as given in his report to Controller Coler in 1900. The data regarding the proposed new Water Supply from Dutchess County and the Catskill Mountains are based on the Burr-Hering-Freeman Report, 1903. Various other data of value have been derived from these reports, and from the report of the New York Merchants' Association on the Water Supply of New York, the special reports of I. M. De Varona, Chief Engineer, on water waste in Brooklyn; N. S. Hill, former Chief Engineer of the Department of Water Supply, on water waste in Manhattan and the Bronx, and various other works, a list of which is given in Appendix E. I also desire to record my appreciation of the courtesies shown me by the Commissioner of Water Supply, Gas and Electricity, Mr. W. B. Ellison, and to the Chief Engineer, Mr. I. M. De Varona, by whom valuable data and suggestions were offered. Respectfully submitted, JAMES H. FUERTES. 103 'I APPENDICES TO WASTE OF WATER IN NEW YORK AND ITS REDUCTION BY METERS AND INSPECTION APPENDIX A: APPENDIX B: APPENDIX C: APPENDIX D: APPENDIX E: UNACCOUNTED FOR WATER WASTE OF METERED WATER WASTE FROM STREET MAINS WASTE REDUCTION IN AMERICAN AND GERMAN CITIES LIST OF REFERENCES I I APPENDIX A. UNACCO UNTED-FOR WATER. T is not known to-day and probably never will be known what becomes of about one-fourth of the water entering the Croton Aqueduct for the supply of New York City. Therefore the statement that half, or more than half, the water coming to New York is wasted is unfounded and misleading. That about half of it has not been accounted for is reasonable; but that half of it is wasted is not true. If all the water coming to the City were measured daily by the most accurate practicable means just before reaching the distribution system, and all water used for public, private, manufacturing and commercial purposes were sold by meter measurement, it would still be impossible to account, year in and year out, for more than three-fourths of the total supply. In general terms, therefore, under the most accurate systems of measurement practicable for such large quantities of water sold in such small units, there is an average of at least 75,000,000 gallons of water per day coming into New York City the ultimate disposition of which cannot be ascertained; and this amount must be deducted from the total reported supply before considering how much wastage can be controlled or prevented. Of this 75,000,000 gallons, one part is made up of innumerable small leaks through imperfect'pipe joints, through the walls of cracked or faulty pipes, through rusted, split or defective service pipes, through the worn-out packing around valve stems, through the faulty and worn-out stop and corporation cocks, and through drip valves at the bottoms of fire hydrants, most of these leaks being too small to be located by any ordinary or practicable means; a second part is stolen through unauthorized or unknown connections and the balance is used but not accounted for by the meters, owing to the fact that many do not register correctly, being worn out or partially clogged. There seems to be a very prevalent idea that every drop of 107 Waste of Water and Its Reduction. water that comes into New York City should be accounted for; and it may be well to indicate briefly why this cannot be done and to illustrate by actual experience in many cities where all water is sold by meter the fallacy of such a supposition. The quantity of water supplied to New York City is estimated by the Engineers of the Department of Water Supply from the depth at which the water flows in the aqueduct at a certain point near the source of supply. At any given cross-section in the aqueduct the water moves with different velocities at different points, the greatest velocities being near the centre of the mass and nearer the top than the bottom, while toward the sides and the bottom the velocities are less. At the place where the depths are read, for determining the discharge, numbers of velocities at different points in the cross section were obtained by means of a current meter, and from these, knowing the area, the mean velocity was determined, from which the discharges corresponding to the different depths of flow were calculated. This method of measurement, under the circumstances, is the only one practicable, and if carefully done should give as accurate a measure of the discharge as any method of which we have knowledge to-day. If the condition of the interior of the aqueduct as to cleanness and freedom from deposits and plant growths remains always the same there is no reason why the discharge for given depths should vary by any considerable amount. This is not the case, however; the conditions are constantly changing and nearly always in a manner to cause the actual flows to be less than those determined when the aqueduct was new and clean. Therefore statements of flow based on the original gaugings are more apt to be too large than too small. The Sudbury Aqueduct in Boston, which is about the size of the old Croton, as well as the Wachusett Aqueduct, which is somewhat smaller than the new Croton, have each shown a reduction of 10 per cent. in discharging capacity from these causes in a single year. The new Croton Aqueduct has been in use almost 15 years and has never been cleaned since first put in service. That its present capacity should be as great as was indicated by the original gaugings is beyond belief; but as no recent gaugings 108 Appendix A: Unaccounted-for Water. have been published, we have now no data by which to judge of its present actual discharging capacity for different depths of flow. The quantity of water used in New York every day is so vast that small errors in estimating the flow reach high figures in the totals. Thus, 309,000,000 gallons, the approximate average quantity of water used per day in Manhattan during 1904, accurately measured by weight on a very cold winter day, would make about one million gallons more than this on a very hot summer day. An error of 10 per cent. in estimating the discharging capacity of the aqueduct would represent an error of 30,000,000 gallons per day in the reported consumption of the City. In explanation of the opening statement in this division of the Report, to the effect that it will probably never be possible to account for more than three-fourths of the quantity of water supposed to be supplied to the City, the data given in Table 12 will be instructive. This table contains the abridged records of 20 American and 17 German cities in which practically all the water used by consumers is sold by meter and for which, therefore, it is possible to make up a statement of the total quantities supplied and total quantities used; the difference between these two quantities represents the unaccounted for water, which includes leakage from street mains and service pipes and all uncertainties in measurement due to inaccuracies in meter measurements and estimates of quantities supplied. On a critical study of these records it will be seen that in most of the cities where the percentage not accounted for is high the water is supplied by pumpage; in these the quantity supplied was estimated from the rate at which the pumps were operated. In some cases arbitrary corrections have been made for the slippage of the pumps and under-registration of the meters through which the water was sold, while in others no allowances have been made. Under actual conditions of operation the amount of water delivered by pumps is generally very much over-estimated when based on revolutions and plunger displacement. The most illumi109 Waste of Water and Its Reduction. nating evidence of this comes from Philadelphia, where all the water is pumped by high duty, well built pumping engines of large capacity, and of a type such that slippage would be supposed to be low. Tests recently made at the instigation of Major Gillette, Chief Engineer of the Bureau of Filtration, showed an actual discharge in the different engines varying from 8 to 56 per cent. less than that calculated by the ordinary methods, and averaging 25 per cent. less for all the large pumping stations from which the city is supplied, their aggregate capacity being over 300,000,000 gallons per day. This explains where some of the unaccounted for water is in the various cities of which the records are given in the table. But making all reasonable allowances for this error, it will be seen that in none of the American cities can over 80 per cent. of the water be accounted for year after year. Perhaps the most interesting and conclusive evidence of the uncertainties atttending the measurement of moving water is to be found in the case of West Orange. The West Orange Water Company buys all its water by meter from another water company and sells it all by meter to the various consumers. Mr. Carroll Ph. Bassett, M. Am. Soc. C. E., the President of the company, writes that they are never able to account for more than 80 per cent. of the water, although by rigid inspections and immediate stoppage of all leaks discovered they are able to keep pretty close to that limit. This is a small supply, distributed through about 30 miles of street mains, and it is to everybody's interest to avoid wastage, yet even under these favorable conditions 20 per cent. of the water cannot be accounted for. All the cities the records of which are given in the foregoing table point to the same state of affairs, and with a uniformity, when local conditions in each are taken into account, which leads to the conclusion that in a city's supply, as an entity, perfection in the measurement of the water, perfection in the stoppage of leaks and perfection in the determination of all the water used are unattainable. In view of all the facts I do not believe that with all preventable leakage stopped in New York, and with all water sold or used through meters, it will ever be possible to account for over 80 per cent. of the water supposed to be supplied; 110 Appendix A: Unaccounted-for Water. and that to account for even 75 per cent. of it would require the maintenance of a skillful, efficient oversight over all meters, to test them occasionally for accuracy, and systematic examinations of the distribution system, to locate and stop underground leaks, combined with periodical house-to-house inspections to discover water connections of which the Water Department has no record. 111 i I APPENDIX B. WASTE OF METERED WATER. AT the present time the water sold by meter measurement in — cludes a large proportion of that used by manufactories, the large stores and office buildings, as well as the hotels and the principal commercial houses. The unmetered water includes that used for domestic purposes, a certain proportion taken by shipping along the water fronts and a few other uses provided for by frontage or assessment rates. These metered consumers, therefore, if now wasting water, are doing so at their own expense and not at the City's; and they will continue to do so at about the same rate. There is one aspect of the question, however, that should receive careful and intelligent attention, and that is, to ascertain whether these consumers are all paying for as much water as they actually take. The City does not own the meters and does not remove them for testing unless a noticeable reduction in the quantity used, as compared with previous experience, is noted in the readings; and it is more than probable that many which have been in use several years and have become worn out do not record correctly. It is also certain that a large amount of water is taken through connections of which the Department of Water Supply has no record. A comprehensive and thorough inspection would disclose these and would add to the revenues of the City. Such an inspection in Philadelphia, in 1900-1901, costing for all clerks, inspectors and incidental expenses $26,447.76, resulted in the discovery of over 70,600 takings of water through fixtures of which there was no record and for which the City thereafter collected about $147,000 a year. There is little reason to think that in this respect conditions are relatively much better in New York than in Philadelphia. In regard to the under-registration of meters in New York a most interesting discovery was made by Mr. E. S. Cole when working under the direction of Mr. N. S. Hill, at that time Chief 118 Waste of Water and Its Reduction. Engineer of the Department, in an extended investigation of water waste, the descriptions of which are contained in the Annual Report of the Department for 1903. During a test of one of the districts then being examined a very large consumption was discovered in Section No. 5, which included the Ansonia Hotel at 73d street and Broadway. Tests were made of the flow in this district from Dec. 6th to 9th, 1903, to locate if possible the source of the heavy draft, with the following results: The meter readings of the Department showed an average daily consumption by the Ansonia of about 53,000 gallons from early in September until the 7th of December. Special meter readings of these meters from December 7th to 9th, by the inspectors looking for leakage, showed an average daily draft at the Ansonia of 239,000 gallons, while the independent measurement of the water flowing in the mains during the same period was 269,000 gallons daily; this included a few residences on 73d and 74th streets in addition to the Ansonia. The rate of night consumption on December 9th by the meters was 173,000 gallons and by the pitometer measurements 200,000 gallons (including the few extra residences above noted), whereas the night rate of consumption by meter was but 7,000 gallons on September 24th, as compared with 137,000 for this same district on this same night by pitometer measurement. It was reported by one of the inspectors that the hood had been removed from one of the large meters prior to December 6th, and the clockwork displaced so as to prevent registration. Examples of this sort can be quoted in great number from the records of other cities, but it is unnecessary to more than point out one such specific instance to show the necessity of constant supervision of all meters if the City is to be able to account for its water and collect a fair amount in return for the water taken. I wish also to point out in this connection that unless the meters are systematically taken out for testing occasionally there is a large temptation set before dishonest employees to allow meters which under-register to stay in place for a consideration. It is practically impossible for the head of the Department to be personally familiar with the behavior of every meter, or to keep 1u Appendix B: Waste of Metered Water. track of the rate of consumption in every building; and unless the meters are tested systematically, sealed and reset there is no possible check on the honesty of the inspectors, upon whom the City now depends for the reporting of defective meters. And as a logical inference, there is no possible way by which the Commissioner can be assured that the City is receiving pay for all the water used through the meters. Another fruitful field for investigation in buildings taking water through meters relates to the surreptitious use of water through connections with the fire protection systems in the buildings. These connections cannot, under a ruling of the Board of Fire Underwriters, be metered, and the taking of water in this way therefore cannot be prevented without frequent inspections, at irregular and unexpected times, by trained men familiar with piping connections and fully alive and awake to the ingenious schemes resorted to by people capable of stooping to such thievery. It is very simple and easy, when pipes come into a building from the street, to tap the pipe outside the building, beyond the meter, carry a branch down and then in again under the floor, and up in some concealed place; and there are doubtless as many men willing to do such work for a consideration as there are men capable of thinking of it. To detect and expose such trickery requires not only honesty, but experience and considerable mechanical training. 116 APPENDIX C. LEAKAGE FROM STREET MAINS. VARIOUS estimates have been made in the past of the probable or possible amount of leakage from the mains and service pipes in the streets of New York; some have been conservative and some wild guesses based on nothing more substantial than the frequently observed fact that in nearly all cities from one-fourth to one-half the water cannot be accounted for, and that a considerable amount of water is apparently used during the night time. From these facts it has been concluded that the water not accounted for must leak out of the pipes, a conclusion which is not conclusive. That there is leakage is a fact, but that all the unaccounted-for water represents leakage and waste is not true. No city has been able yet to determine the exact amount of leakage from its mains, as the possible means of making such determinations are in themselves but approximate. It is manifestly impossible to make such tests in any large city without jeopardizing important interests. One of the prime requisites for obtaining data on this important subject is that all service pipes be shut off when the tests are made, otherwise the amount of water found running through the pipes at any point would represent use or wastage in the buildings, not leakage from the mains. In New York such tests as this could not be made, as there are no districts where all service pipes are provided with shut-off cocks; In fact there are very few such cocks outside the buildings. Even if stop cocks should be provided on every service pipe it would still be impossible to make tests in certain sections of the City where the legitimate night use is almost as great as the day use. Serious leaks can generally be readily found; but the innumerable small ones are costly to locate and therefore costly to stop, not only in money but in time. The location of the more important of the general leaks can 117 Waste of Water and Its Reduction. be determined by a detailed examination of the flow of the water in the mains, street by street, during the night time when the general use of water is the least. An abnormal rate of consumption of water in a certain section would then indicate excessive use, excessive wastage or excessive leakage, and a hunt through the premises and sewers in the section under examination would account for either the excessive use or waste. If neither excessive use nor waste is discovered a more detailed examination of a smaller section will locate the leakage from the mains and service pipes. It may readily be seen that such an examination would involve years of systematic work on an elaborate scale to stop a significant portion of the leakage in the 850 miles of street mains and probably more than 850 miles of service pipes in Manhattan alone. For immediate relief, therefore, no hope is to be held out that any considerable reduction of wastage can be accomplished in this direction. Nevertheless work of this character should be undertaken immediately and continued as a part of the permanent routine work of the Department. Mr. I. M. De Varona, Chief Engineer of the Department of Water Supply, is now arranging for a systematic house-to-house inspection of all properties in New York in conjunction with examinations for leakage, not only in the buildings, but in the street mains as well. The plan outlined by Mr. De Varona contemplates the control of wastage in the fixtures in buildings by the imposition of fines for all leaks found on second inspection. The leaks are to be located by house-to-house examinations combined with tests of the flow in the mains, during the night time, coincidently with examinations of the flow from the sewer connections in the same district. The tests of the flow in the mains will be made by experienced observers with pitometers to be owned by the City, through permanent fixtures to be attached to the mains. This system will, if carried out extensively and maintained as a permanent branch of the organization, and with capable men in charge, control a large portion of the wastage. I venture to predict, further, that if carried out thoroughly the increased annual revenue resulting from the discovery of water takings not recorded on the City's books at the present time will more than pay 118 Appendix C: Leakage from Street Mains. the cost of the first year's inspection as well as for the equipment of the bureau. Just as soon as the system is discontinued, however, through changes of administration, withholding of appropriations, or other exigencies of a political nature, the benefits derived from it will be immediately lost, and the former conditions of wastefulness will return. While there is considerable information to be had regarding the tightness of pipe lines, as experimentally determined, such data can not be applied to conditions in a City where valves, hydrants, corporation cocks and service pipes, some of them very old and much rusted out, are incorporated in the distribution system in endless numbers; and it seems futile to quote such figures and attempt to apply them to New York conditions. We have no way of separating this leakage from the rest of the unaccountedfor water, part of which is stolen and part of which can not be accounted for because there are always many meters in use which do not register correctly, either because they have been tampered with, or have become worn out, or clogged with rust, or obstructed with vegetal fibres, or fish, or eels, or frogs; and even with the closest inspection and most frequent testing practicable all the meters can not be kept in order all the time. As a matter of fact, therefore, the only really practical indication of excessive street main leakage is to be found in the amount of unaccounted-for water. Where this is found to be relatively no more than in completely metered cities using every precaution to keep wastage and leakage stopped it may be assumed that there is nothing serious the matter with the distribution system taken as a whole. This is the case in both Manhattan and Brooklyn. Further confirmation of this position is afforded by the examination of the water mains uncovered during the construction of the subway system, and which I, personally, have taken occasion to observe many times, at many places, in both Manhattan and Brooklyn. Still further confirmation is to be had from the extensive tests made in 1903 under the direction of Mr. N. S. Hill, then Chief Engineer of the Water Department, covering about 12 per cent of the entire area of Manhattan Island, and by Mr. I. M. De Varona, at that time Chief Engineer of the Brooklyn Water Supply in Brooklyn. 119 Waste of Water and Its Reduction. These tests were very carefully and thoroughly made; and yet they failed to locate any significant leakage from the mains. In the downtown office building district where it was surmised that leakage would be discovered, if anywhere, the pipes being among the oldest in use in the City, it was found in the district tested in detail, including Liberty street from Greenwich to West street, and West street from Cortlandt to Cedar street, that not only was there no evident leakage but that there was practically no flow observable in the mains after midnight. The rate at which water is used during very early hours of the morning is very frequently taken as a standard for indicating the prevalence of wastage and of leakage from mains, but such data must be used with judgment and with a thorough knowledge of the various modifying conditions. The sweeping statement that the water found moving through the mains between 2 and 4 A.M. indicates corresponding leakage and waste is only true when it can be proved that no water was used during that period, and that the valves in the mains surrounding the district under test were closed, and were tight. It is, no doubt, true that if there were no use of water during the night and no leakage through fixtures in the buildings, the motion of the water in the mains would indicate leakage; but on the other hand the finding of a night consumption does not by any means prove leakage and wastage. The City of Bayonne buys all its water by meter measurement and sells it to consumers by meter measurement. Just below Bayonne, on Constable Hook, are the large works of the Tide Water Oil Company, the Orford Copper and Sulphur Company, Pacific Borax Company, and Columbia Oil Company, with numerous docks, boilers and pump houses, chemical houses, warehouses, paraffine works, and other allied industries, all of which are supplied with water, through meters, from the Bayonne mains. On the 13th and 14th of October, 1899, under the writer's direction, hourly readings were taken of the meters through which the entire supply to the city passed and of all the 20 meters on the service pipes supplying the various plants on Constable Hook. The general results for that 24 hours were as follows, based on the resident population of Bayonne which at that time was about 21,000: 120 Appendix C: Leakage from Street Mains. Average daily con- Minimum rate of sumption per cap- night consump- Per ita, gallons. tion per capita, centage gallons. Entire City, including Constable Hook... 161 102 63 Constable Hook alone..81 58 71 Bayonne, (excluding Constable Hook)... 80 16 20 In the entire city, including the Hook, the maximum rate of consumption, 223 gallons per capita per day, was reached at two o'clock in the afternoon and the minimum rate, 102 gallons, at four o'clock in the morning. In the Hook alone the maximum rate, 105 gallons per capita, was reached at midnight and the minimum, 58 gallons, at 9 o'clock in the evening, while in the city alone, the maximum rate, 150 gallons per capita, occurred at 4 o'clock in the afternoon and the minimum, 16 gallons, at 6 o'clock in the morning. During the night time when the meters were being read in the Hook no visible operations were going on, the entire district was as dark as pitch, being practically deserted except for a few watchmen and men stationed at certain points to keep up steam and keep the pumps running, yet the amount of water actually consumed, as shown by meter measurements, was but little less during the night-time than during the day, and the actual maximum rate of consumption was at midnight, at which time it would naturally be supposed that comparatively little water would be used. Deducting the water used in the Hook from the total amount entering the Bayonne mains, however, we obtain the characteristic conditions found in residential cities: maximum rates at 9 to 10 o'clock in the morning and 2 o'clock in the afternoon, with minimum rates in the early morning hours. An examination of the individual records of the meters in the Hook showed the interesting fact that some plants used large quantities of water at a comparatively uniform rate throughout the entire 24 hours, while others used almost none for an hour or two at a time, and then large amounts at irregular intervals. 121 Waste of Water and Its Reduction. This instance is quoted at some 'length to show how very far general conclusions based on only theoretical considerations may vary from actual practical conditions and to indicate the danger of concluding that if tests show a minimum night rate equal to a considerable proportion of the average daily rate of consumption that this excessive night rate points only to excessive leakage and wastage. New York is the most important commercial and manufacturing city on the continent, with industries of various kinds using water at night in large quantities, and a relatively high night rate of consumption is to be expected; this night rate, therefore, without definite information, section by section, regarding conditions all over the city will bear no definite relation either to wastage or leakage and will be no standard by which the amount of either can be estimated or determined. In some sections of New York City the pressure in the mains is frequently not great enough during the day time to put the water in the upper stories of the buildings, and tanks, placed at such an elevation that the water will rise up and fill them during the night, are extensively used. In a very considerable proportion of the low-level district the combined capacity of these storage tanks is nearly half the total amount of water used daily in the districts; and to fill them all, under the dribbling streams coming from the slowly rising pressures as the consumption decreases towards the end of the day, may well consume many hours at night. In 1900 the writer had occasion to remodel the water supply arrangements in a high-class apartment house located on Madison avenue above 60th street where the pressures in the street mains were generally so low during the morning that the water would not rise above the second story of the building. The building contained several apartments renting at the average price of $10,000 per year each and in order to keep them occupied an adequate supply of water was necessary. Each apartment was lavishly provided with sinks, baths, toilets, bidets, showers, etc., with a general laundry and a refrigerating plant in the basement for the manufacture of ice not alone in the basement but in the private refrigerator of each apartment. The remodelled plant provided 122 ' DIAGRAM NO.2 SHOWING HOURLY VARIATION IN RATE OF CONSUMPTION DURING SEVERAL DAYS IN BAYONNE, BROOKLYN, CHICAGO, DETROIT, FALL RIVER, HARTFORD, NEW YORK, PROVIDENCE AND WOONSOCKET I 0 I 0 I 0 I 0 I 0 I o c i 0o 0 0 0 0C 0 0 0 0 0 0 o 3 Z Z Z z Z Z z N1 _ N _ NN N N NC Bayonne shown thus Curve of Bayonne, from observations made under supervision of J.H.F. Brooklyn.~.. Curves of Brooklyn, Fall River, New York and Woonsocket from Report of John R.Freeman. Chicago.,,t Curve of Chicago, from Report by John Ericson, M.Am.Soc.C.E., City Engineer, May 1905, on Water Supply System Detroit i;, _of Chicago: its Past, Present and Future. Fall River it +++ ++ Curve of Detroit, from Annual Report of Board of Water Commissioners, 1898. Hartford. _+_._ —+ Curve of Hartford, from Annual Report, Board of Water Commissioners, 1904, New York., Curve of Providence deduced from observations made June 23-26, 1891 under supervision of J.H.Shedd. C.E. Providence " All Lines showing Rate of Consumption drawn to same scale. Woonsocket " __ Appendix C: Leakage from Street Mains. a storage tank in the basement with a capacity equal to half a day's total supply and two tanks on the roof holding, together, a half day's supply. This large provision was made as a safeguard against an interruption of the supply due to breaks in the mains, which had been frequent in that vicinity, and had generally not been repaired in less than 24 hours. All through this district the various buildings have pumps operated by steam, electricity, or hot air engines in the basements, with storage tanks in the attics; ia some the pumping is done only in the day-time, while in others the pumps run practically all the time. In the large building above referred to the pumps are small and are controlled by an automatic device. During the morning, middle of the afternoon and early evening the heavy drafts from the roof tanks lower the water for short periods much faster than the pumps can supply it, with the net result that the pumps run at intervals all through the day and night. In 1903 Mr. Hill found a night consumption of 106 gallons per capita in a district including 250 acres in the Bronx. The mains in this district were new. Sewer inspections and other tests revealed no main leakage and but a very small amount of fixture leakage in the 44 unmetered buildings in the district. Most of the water supplied to this district was used for manufacturing purposes. Serious error is likely to result from the assumption that the minimum night rate of consumption in a large city represents or is indicative of the extent of waste and leakage. In Diagram No. 2 the average daily rates of consumption per capita, in several cities, for each hour during several consecutive days are arranged for convenient comparisons. From the Chicago diagram the ordinary night rate was about 165 gallons per capita and the ordinary maximum day rate about 195 gallons per capita, the difference, if the night consumption represented waste and leakage, 30 gallons, would then represent the actual amount of water necessary. In Providence, on the other hand, the ordinary night rate was about 20 gallons per capita and the heavy day rate about 85 gallons, leaving 65 gallons per capita per day 128 Waste of Water and Its Reduction. to represent the actual water used. According to this, Chicago, a live, wide-awake, progressive, rapidly growing city, with all its enormous manufacturing, shipping, railroad, commercial, professional, packing house, trades, and public uses, ought, theoretically, if only the leakage and wastage could be stopped, to get along with a per capita allowance of water less than half that required in Providence. It is very evident, from a comparison of the lines on this diagram, that the night rate of consumption indicates more regarding the character of the manufacturing and commercial industries in a city than it does of the amount of wastage or leakage; and that it can not be a measure of the amount of wastage and leakage unless the actual night requirements for manufacturing, commercial and industrial uses be determined; and this, in a large city like New York, even if completely metered, would be not only impracticable but next to impossible. It appears to me that data concerning night consumption in large cities are really of little practical value for the use to which they most often are put: that of forming a foundation on which to build up demonstrations of excessive wastage. In small, restricted districts where exact conditions are ascertainable, their application to facilitate the location of wastage and leakage are proper, and in strictly residential cities they may give an approximate indication of the extent of such losses. In any event it is apparent that, so far as New York City is concerned, it is hopeless to expect that such leakage as may occur in mains or service pipes outside the houses could be reduced in sufficient measure to appreciably affect the general rate of consumption short of many years of patient, systematic and very costly work; and to harp on this source of preventable waste as one which should be immediately attacked as a means of relieving a shortage of water in New York City is nonsense. In about ten or fifteen or perhaps twenty years of such work it might be possible to save, perhaps, 5 to 10 gallons per capita of the water now leaking away in this manner. No doubt this water is worth saving when detected in conjunction with the reduction of such other and more important wastage as can be found immediately and stopped. The proper way, therefore, to handle this street main 124 Appendix C: Leakage from Street Mains. leakage is to look out for it when searching for the more serious and extensive wastage and leakage going on in premises where no meters have been installed, and in the hunt for illicit takings of water. 125 APPENDIX D. WASTE REDUCTION IN AMERICAN AND GERMAN CITIES. PART I. AMERICAN CITIES. ASBURY PARK, NEW JERSEY. THE Annual Reports of the Water Department of Asbury Park are not published except in the local papers, and hence extensive data are not to be had. The following information has been kindly furnished by Mr. John L. Coffin, Superintendent, under date of March 7th, 1906. Asbury Park is a summer resort whose "busy season" is from June 1st to September 15th, the culmination being about August 15th. Fo.r several years prior to 1899 there had been August shortages of water that were annoying, but in 1899 the shortages were so great that the pressure on the mains during the day was below ten pounds, and a neighboring plant was called on for assistance. It had become necessary to stop all lawn and street sprinkling, and the prosperity of the city was seriously threatened. After the close of that season Mr. Coffin began a systematic campaign for the adoption of meters. There was great opposition but the famine of the past season was fresh in the minds of the Council, and the measure was finally passed in April, 1900. Half the city was metered in that Summer and the season passed without a shortage; by July 1, 1901, every service in the city was metered, and during the five years following, without increasing the supply in any way, there has been an abundance of water at all times. In addition there has been furnished an average of 200,000 gallons 127 Waste of Water and Its Reduction. per day for street sprinkling, or more than twice the quantity ever before allowed for that purpose. The growth of the city in the last two years has been unprecedented, and it has become necessary to increase the supply. But the meters held their place as checks on waste and the vast majority of property holders would not return to the old "flat rate" system. The receipts of the Department have increased from $20,000 in 1899 to over $26,000 in 1905, while the rate has been reduced from $2.50 to $1.50 in the same period. The city furnishes and takes care of all meters, the property owners having to install the lead pipe connections and stop valves. A quarterly minimum charge of $1.50 is made on each meter connection, whether water is used or not, and this charge entitles the consumer to 1,000 cubic feet. No accounts are kept with tenants. The plant being under municipal ownership, the charge is by law against the property, arrears of water charges being a lien. The total consumption of water at Asbury Park during the past seven years was as follows: 1899 183,000,000............. Flat rates. 1900 188,000,000...................About half the services were metered by the close of the season. 1901 152,000,000...................All services metered. 1902 168,000,000.................. " " " 1903 195,000,000................. " " " 1904 206,000,000................... " " " 1905 231,000,000..................." " " There are no manufacturing interests at Asbury Park, but about 15 per cent. of the total consumption is used for sprinkling streets, washing filters and other public purposes. The reduction of the consumption in 1901, the first season when all services were metered, amounted to about 20 per cent. of the previous year's consumption, at the close of which about half the water was being sold by measurement. The probable reduction is estimated by Mr. Coffin to be about 45,000,000 gallons, or, counting in the natural increase checked by the meters in 1900, a total probable reduction of 24 per cent. Since 1900 there has been a continual and steady increase in consumption, the large increase during 1905 being attributed to the great conventions of 128 Appendix D: Waste Reduction in American Cities. early July, the effects of which were felt throughout the season. At the present time there are no restrictions as to the use of water, and bouse-to-house inspections have been abandoned. The price of water is the same to all consumers and for all quantities, meters being read and bills rendered quarterly, with two additional readings for August 1st and September 1st. The losses from leakage from mains and service pipes cannot be very great, as frequent tests show that in Summer the rate of consumption between 6 P. M. and 6 A. M. is about 30 per cent. of the total day's rate; while in Winter, during the same hours, it is about 20 per cent. ATLANTA, GEORGIA. IN Atlanta, Georgia, the consumption of water had become so great in 1884 that it was necessary to put in an extra pump in order to keep up the supply and maintain the necessary fire pressure. After this pump was put in operation it soon became evident that another force main would be required, owing to the great loss of pressure resulting from the excessive taking of water from the mains; and as it was thought impossible to raise the money for this purpose, great objections being made to the increase of taxation, and as the insurance companies complained of the insufficiency of fire pressures, it was determined to try to reduce the waste of water by selling it to consumers at meter rates instead of flat rates. The meters were placed, controlled and kept in repair by the Water Commissioners, but were paid for by the consumers, though much complaint was made at the time that it was unjust for the consumers to have to pay for their meters and then pay for the water measured through them. This policy was adopted for the reason that but one-third of the city was within reach of the mains. To tax the other twothirds for the meters installed for the control of the one-third using the water, which one-third would also enjoy the benefits of fire protection and decreased insurance rates, appeared unreasonable; and there was no way for the city to get the money for the purchase of meters except by taxation. 129 Waste of Water and Its Reduction. The general introduction of meters began in 1885, although a few had been installed in previous years. In 1897 practically all the services were metered, and since that year metering has been compulsory. The records of Atlanta therefore are of particular interest. It is a completely metered city, selling water at a high rate per 1,000 gallons and with a comparatively high per capita consumption. Unfortunately, no census of the consumers has been taken, and there is therefore no direct comparison possible with the other cities where such data have been kept. The introduction of meters put off for a time the necessity of increasing the capacity of the plant, but by 1892 the population had nearly doubled and a new water supply system with a source of supply sufficient for a long time in the future became a necessity. Accordingly, new works were built and put in operation in 1893; an additional 15,000,000 gallon pump was added in 1901, and still, at the present time, the plant is taxed to the limit of its capacity, notwithstanding the fact that all services are metered and the charges for water high. The average daily pumpage of water, the number of services, the progress of metering and the average daily consumption per service from 1883 to 1904 inclusive are given in Table 24, (see opposite page). The cost of the Meter Department, exclusive of the cost of reading the meters, from 1899 to 1904, has been as follows: TABLE NO. 25. Cost of De- Amount ColCost of Number partment, Ex- lected from Con- Net Cost to YEAR. Meters of Meters clusive of Cost sumers for Water DepartSet. Set. of Meters. Parts of Meters ment of OperatReplaced. ing Meter Department. 1899 $11,067.70 935 $3,584.16 $1,353.40 $2,230.76 1900 11,358.50 821 3,837.35 1,264.30 2,573.05 1901 10,667.30 921 3,426.64 749.60 2,677.64 1902 20,084.20 1,539 3,558.32 815.95 2,742.37 1903 18,138.50 '1,520 4,169.63 724.60 3,445.03 1904 24,450.70 2,142 3,050.90 105.80 2,945.10 Average cost, per meter'set, $12.15. 330 Appendix D: Waste Reduction in American Cities. TABLE NO. 24. Average Total Annual Number Daily Con-, o YEAR. Pumpage, of sumption Cs a Remarks. Gallons. Services. per Service, - Gallons. ~'i 1883 859,448,080 1,663 1,410 183 95 meters. 1884 1,148,420,700 1,832 1,720 225 570 meters. 1885 876,055,000 1,973 1,220 160 General metering. 1886 574,925,000 2,077 820 107 1,372 meters. 1887 No data 2,192... From this year, 1888 628,931,300 2,464 700 91 practically all 1889 756,762,600 2,883 720 94 services are 1890 861,241,100 3,273 725 95 metered. 1891 942,966,730 3,759 698 91 1892 923,451,700 4,533 560 73 1893 1,296,550,800 4,934 720 94 1894 No data 5,544 1895 1,657,549,110 5,962 765 100 1896 1,814,963,500 6,496 715 93 1897 1,895,623,800 7,176 720 94 1898 1,807,661,800 7,974 620 81 1899 1,996,908,200 8,592 638 83 1900 2,146,635,700 9,277 634 83 1901 2,261,947,100 9,926 624 81 1902 2,511,687,200 10,676 642 84 1903 2,484,954,300 11,654 584 76 1904 2,761,337,800 13,001 582 76 *These figures are based on 7.65 persons per service, and are given merely for purposes of comparison. Average consumption per service from 1888 to 1904, inclusive, 677 gallons per day. Minimum, 84 per cent. of average. Maximum, 115 per cent. of average. 181 Waste of Water and Its Reduction. The consumers pay for meters and for the renewal of worn-out parts, the city selling the meters to the consumers and making the necessary repairs when needed. The total cost of the maintenance of the meter system, exclusive of the cost of reading the meters, sending out bills, etc., but including the replacement of condemned, worn-out meters, with new ones, is from four to five per cent. per annum on the value of the meters in service. This is on the assumption that all the cost would be borne by the city, as would be the case if the city purchased, owned and maintained the meters. It is interesting to note the gradual and progressive reduction of consumption per service during 1885 and 1886, while the metering of the old services was in progress, and still more so to note the almost constant rate which has prevailed since the city has been completely metered. The average rate from 1888 to 1904 inclusive has been 667 gallons per service, the minimum average consumption for any year being 84 per cent. and the maximum 115 per cent. of the mean. Based on 7.65 persons per service, which is the ratio for 1889 (deduced from a population of about 65,000 persons supplied with water), the average daily consumption for these 17 years has been about 85 gallons per capita, with a tendency to decrease rather than to increase. This record is old enough to be entitled to respect and demonstrates that in a city on a permanent meter basis the question of waste is eliminated from consideration. No house-to-house inspections have been made or are necessary. The ordinary situation is reversed; it is profitable to the Atlanta Water Department to have water wasted and unprofitable to the consumer to waste it. On the other hand, there seems to be no excessive economy in the use of water, as 85 gallons per capita is a rather high rate of consumption in a city of 100,000 people where all the water is filtered and sold at meter rates at 20 cents per 1,000 gallons. SUMMARY.-The important facts illuminated by the Atlanta experience are the following: First.-The metering of all consumers, manufacturing, domestic and commercial, was followed by the reduction of the con132 Appendix D: Waste Reduction in American Cities. sumption from about 200 gallons per capita to an average of about 85 gallons per capita daily. Second.-The benefit of the system has been permanent for 17 years and presumably will remain so as long as universal metering is practised. Third.-With the complete metering of the city the rate of consumption per unit of population supplied has remained at a nearly constant figure, despite the rapid growth of the city in population, manufacturing interests and commercial importance. Fourth.-The placing of meters has not been followed by the reduction of the per capita consumption to the exceedingly low figures obtaining in many European and in a few American cities. ATLANTIC CITY, NEW JERSEY. A TLANTIC CITY, a resort on the New Jersey coast, has a resident population of about 28,000 during the fall, winter and spring months, while during the three summer months from 100,000 to 150,000 guests are taken care of in the many large hotels, boarding houses and cottages. This water plant therefore has a very unusual condition to meet,the extremes between the minimum and maximum amounts of water required being much greater than in ordinary water supply works, the ratio between the day of smallest and the day of greatest consumption being often as great as 1 to 6, instead of the more usual ratio of 1 to 2. The amount of reserve pumping capacity required, therefore, is necessarily greater than usual and the difficulties of foreseeing maximum requirements correspondingly greater. In 1895-6 the works were taxed to their utmost to keep up the supply, the water being furnished 'a discretion; and at times the draft on the system was so heavy that there was no pressure on the mains in certain parts of the area supplied. There were, on the 1st of August, 1895, but about 350 meters in use, and no methods were in force by which to restrain excessive wastefulness. The situation was so critical that immediate action was necessary; without an adequate supply of water the place would lose its desirability as 133 Waste of Water and Its Reduction. a summer resort; and the extensive, profitable investments be rendered worthless. To meet this situation the Board of Water Commissioners, who estimated that fully half the water pumped was wasted, and that the supply was ample if properly used, decided to change from the system a discretion to the system of sale by measure. Accordingly an appropriation of $25,000 was made for the purchase of meters and during February, 1896, the first lot arrived and placing them was immediately commenced. By the 15th of July some 1,500 meters had been set, which, with the original 350 in use, controlled about half the services. In connection with the placing of the meters a house-to-house inspection was started which revealed many hundreds of leaky service pipes and fixtures, and disclosed many fixtures of which the Department had no record and on which a considerable revenue has since been collected. The following tabulation contains the estimated resident and summer populations, the average daily consumption, number of services and meters in use, percentages of services metered and estimated per capita consumption during the summer months. TABLE NO. 26. ESTIMATED A5 O~ POPULATION. )43, -' Average No.'of o Daily Number Number Servi- P "t YEAR. Con- of of ces m m Summer sumption, Services. Meters. Me- o o Resident. Population, Gallons. tered. U g Average. j - 1 -1895-6 23,000 125,000 5,259,152 3,446 1,806 52 1896-7 24,000 125,000 3,171,321 3,492 2,135 61 1897-8 25,000 125,000 2,294,787 3,689 2,636 71 40 1898-9 26,000 150,000 2,372,457 3,956 3,054 77 35 1899-0 27,500 150,000 3,025,723 4,249 3,298 77 35 1900-1 31,500 150,000 3,412,854 4,632 3,639 78 35 1901-2 35,500 150,000 3,741,176 4,477 3,662 81 35 1902-3 37,500 125,000 3,927,401 4,819 3,929 80 35 1903-4 37,500 125,000 4,878,591 4,903 3,981 81 55 1904-5 37,600 125,000 5,194,339 5,150 4,038 78 58 During the fiscal year ending August 1st, 1896, the introduction of meters was commenced in February and the 1,456 meters 134 Appendix D: Waste Reduction in American Cities. placed were therefore set during the last half of the year, so that the checking of wastage did not affect the first six months' consumption. By the next year, however, with 61 per cent. of the services metered the effect was decided, and during the following year, when 71 per cent. were controlled by meters, the consumption dropped to less than half what it had been years before. It will be seen that by the checking of wastage the average daily consumption has been kept below the point reached in 1895 for nine years. In 1898 the consumption fell to the figures which had prevailed in 1892, but has been steadily rising until 1905 it was about the same as in 1895. The reduction in wastage affected the consumption in May (a month when the visiting population is comparatively small) in the same amount as that in August (when the summer population is greatest), the reduction being about 3,000,000 gallons per day. This therefore is a measure of the loss due to leaky fixtures and poor plumbing, corrected in two years. The reduction amounted to 65 per cent. of the May consumption, and 40 per cent. of the August or summer consumption. Since the time that the low point was reached in 1898 the May consumption has increased by 3,200,000 and the August consumption by 3,900,000 gallons daily, the increase in each case being a steady rise, chargeable partly to increased population and partly to the greater number of fixtures in use and higher standards of living. With such a changeable population it is, of course, impossible to arrive at reliable statements of the per capita daily consumption. From a careful study of the amounts of water pumped each month since 1892, it would appear that the summer population uses about half as much water per capita as the resident population during the remainder of the year, and the amount used by each class of consumers has been increasing each year in a greater ratio than the increase of population. The explanation of this increase is not clear. Possibly there is still considerable wastage, notwithstanding the metering of over three-fourths of the consumers. The population is essentially residential, the amount of water used for public, trades and com135 Waste of Water and Its Reduction. mercial purposes being relatively insignificant in proportion to the total amount. The figures given in the last column of the table are the average per capita consumptions for several years during the three summer months as given in the reports of the Board of Water Commissioners. Assuming these to be approximately correct, the per capita consumption of the residential population during the Fall, Winter and Spring would have been about 65 to 75 gallons per capita, up to 1902, increasing yearly to about 100 gallons in 1905. Some trouble has been experienced with electrolysis, the action being more extensive on service pipes than on mains, numbers of services requiring renewal every year. The cost of purchasing and setting 1,456 meters in 1906, including valves, fittings and labor, was $15.19 per meter. In connection with the installation of meters, Mr. W. C. Hawley, Superintendent, states in his report to the Board of Water Commissioners for 1899: "The meters have saved money, not only for the Department, thus benefiting every taxpayer, but the large majority of takers pay less under the meter rates than they formerly paid under the assessment rate. Those who allow leakage or waste have to foot the bills therefor, and the result is that more care is taken to keep plumbing in repair and to prevent waste. Better plumbing is being put into new buildings and more care is taken to build so that pipes will not be likely to freeze. But few complaints are made concerning the meters now, and strange to say, most of these are because the meters have not registered enough. In other words, the consumers find it impossible to use the whole amount of water to which the minimum rate entitles them, and hence the complaint, in spite of the fact that the minimum rate charged is less than the former assessed rate." When meters were first introduced there was a tendency to practice rigid economy in the use of water, but the consumers have learned that this is unnecessary, as the minimum rate entitles them to an abundance of water for all proper purposes, and the consumption consequently increased again up to a figure near what the minimum rate entitles them to. 136 Appendix D: Waste Reduction in American Cities. SUMMARY.-The introduction of meters, coupled with house-to-house inspections during the first year, resulted in the reduction of wastage to the amount of 55 per cent. of the supply, with 77 per cent. of the services metered. For a number of years there was no material increase in the amount supplied per capita, but since 1903 there has been an upward tendency. The rate per capita in 1905, based on resident population, was not over half the rate in 1896, before restrictive measures were put in force. After the introduction of meters the total average daily consumption did not reach the amount required in 1896 for about nine years, notwithstanding the increase in population and multitudinous new uses which have developed for water in recent years. CLEVELAND, OHIO. NE of the most interesting and successful campaigns for the reduction of wastage was started in Cleveland in April, 1902, under the direction of Mr. E. W. Bemis, Superintendent of the Water Works. During the five years ending in 1901 the annual pumpage of the Department increased 44 per cent., which was over twict as fast as the increase of population of the city; the increase of the use of water in all the factories, hotels, apartment houses and large stores, every one of which was metered throughout that period, was at a somewhat lesser rate. Cleveland, like all growing cities, has been for years facing the problem of increasing consumption. As long ago as 1884 the report of the trustees called attention to the great waste of water evident by the amount consumed, particularly in protracted periods of cold weather in Winter and dry, hot weather in Summer, contrasting the water consumption in Cleveland with that of several Eastern cities in which meters were freely used and showing a condition of wastefulness calling for vigorous measures for its abatement. Table 27 (see following page) shows the estimated population in the city proper and suburban districts supplied from the Cleve137 Waste of Water and Its Reduction. TABLE NO. 27. Average Average Per Per Pop- Average Daily Daily Cent. Cent. ulation Daily Con- Con- Number Number of of YEAR. Supplied Consumption sump- sump- of of Ser- Sup City and of Water, tion per tion per Services. Meters. vices Suburbs.* Gallons. Con- InhaMe- gold sumer, itant, tered. by Gals. Gals. Meter 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 45,000 46,000 47,000 50,000 53,000 54,000 55,000 60,000 65,000 72,000 80,000 85,000 90,000 100,000 105,000 115,000 118,000 124,000 130,000 134,000 138,000 142,000 150,000 156,000 172,000 180,000 185,000 198,000 200,000 214,000 231,000 243,000 251,000 261,000 290,000 310,000 325,000 345,000 348,000 353,000 355,000 375,000 400,000 400,000 411,000 424,000 438,000 455,000 348,664 398,467 513,107 710,984 881,599 1,012,794 1,152,875 1,300,858 1,417,153 1,609,239 1,907,861 2,106,265 2,462,839 3,085,558 3,746,907 4,607,571 5,095,230 5,625,150 6,073,358 6,573,220 7,726,920 7,925,882 9,446,498 10,179,461 13,280,025 12,313,804 14,212,144 16,367,153 17,950,694 19,530,611 22,266,155 23,199,703 24,875,236 27,787,158 32,237,077 36,442,989 42,163,194 39,491,876 47,154,366 45,629,875 48,379,371 51,839,816 61,712,984 67,087,942 69,648,383 69,965,742 62,010,650 61,572,022 111 93 91 105 114 121 117 124 123 124 116 116 120 113 125 132 138 141 137 131 142 135 146 145 162 136 137 138 139 137 140 130 105 129 136 143 158 134 151 142 150 149 169 173 174 173 145 138.5 8 8 11 14 16 19 21 22 22 22 24 25 27 31 36 40 43 45 47 49 56 51 63 65 77 68 76 83 90 91 96 95 98 106 111 117 130 114 137 129 136 138 153 169 169 165 142 136.8 49,832 52,303 53,473 55,130 56,816 58,852 60,627 2,416 2,617 2,805 3,344 11,099 25,030 30,226............ 4.3 5.0 5.2 6.1 19.5 12.5 49.2.22. 23.6 25.0 27.2 35.. 3 39.3.... *Approximate. 138 Appendix D: Waste Reduction in American Cities. land Water Works, the average daily consumption of water and the average daily consumption per inhabitant and per consumer, from 1857 to 1904, inclusive, the number of services in use, the number of meters in use, the percentage of services metered and percentage of total consumption passing through meters from 1894 to 1904, inclusive. This table is very instructive. Comparing the columns giving the average daily consumption per inhabitant and per consumer, it will be seen that in 1857 so few of the properties in the city were taking water that the per capita consumption was only 8 gallons per day, but that those who did use city water used it almost as extensively as at the present time. This points out one of the dangers of comparing the per capita consumption of different cities, or of attempting to judge of a city's needs at the present time by the amount of water used per capita during former years. The amount of water used per consumer has had a general tendency to increase since about 1860, although between that time and 1890 there were several years when it dropped to as low a figure as prevailed in 1857. The first evidences of persistent increase began about 1890 and continued until about the middle of 1902. In 1901 it became imperative to reduce the wastage, and acting under an ordinance passed and approved in August, 1901, giving the Director of Public Works authority at his discretion to place meters on any or all water connections, both business and residence, preparations were immediately made to extend the meter system to domestic consumers as rapidly as possible. The city furnishes, sets and maintains the meters, making all repairs except those arising from frozen meters when set in basements, or meters destroyed by hot water, the expense of repairing which is collected from the consumer. On January 1st, 1902, there were in service 3,344 meters, of which 379 belonged to consumers. All these meters were on large consumers, such as factories, hotels, large stores, apartment houses, etc. Beginning in April, meters were added rapidly and by the end of the year 7,739 public and 16 private meters had been added, making the total number in use by Decem139 Waste of Water and Its Reduction. ber 31st, 1902, 11,099, including all sizes, both public and private. Attention is particularly called to the fact that at the end of 1899, while only 5 per cent. of the services were metered, these few meters were passing about 22.4 per cent. of the total pumpage of the city, the metered water at that time representing practically all the manufacturing, trades, hotel and apartment house consumption. The same condition obtained through 1900, 1901 and the first half of 1902, and the metered water up to that time did not include that taken by any domestic consumer, yet the per capita consumption remained at a very high figure. With the extensive introduction of meters on domestic and small trades consumers, in 1902, an immediate reduction began to take place, but the number of meters reached was not sufficient to show in the yearly average in a very marked degree. During the succeeding year, 1903, the number of meters in use was more than doubled and by the end of 1904, nearly half the services in the city were metered and about 39 per cent. of the total pumpage was sold by measure, the per capita consumption having fallen steadily as wastage was controlled. From 1901 to 1904 the percentage of services metered increased from 6.1 to 49.2, or 43.1 per cent., while the percentage of the total supply sold by measure increased from 25 to 39.3, or 14.3 per cent. This increase was largely on domestic consumption. Since the total pumpage includes, in addition to the water used by all consumers, water used for public purposes, loss by leakage from mains and service pipes, water used at the plant, under registration of meters, slip of pumps and ~waste by unmetered consumers, the 14.3 per cent. increase in total pumpage actually represents the metering of about 39 per cent. of the domestic consumers. Assuming that the extension of the meters to the balance of the domestic consumers should effect a further reduction of wastage in proportion to that so far obtained, there should be attainable a total reduction of 73 gallons per day, giving for a probable reasonable use of water for all purposes in Cleveland 96 gallons per capita daily. The large trades and manufacturing consumers apparently require 140 Appendix D: Waste Reduction in American Cities. about 40 gallons per capita daily, hence the probable attainable conditions in Cleveland might be expressed thus: Gals. Consumption per capita daily in 1901.................... 169 Reduction by metered 39 per cent. of domestic consumers 32 Possible reduction if remainder were metered............ 41 73 Probable attainable average............................. 96 Manufacturing and trades uses.......................... 40 Balance, for domestic consumption......................... 56 Public uses and unaccounted-for water.................... If of this we assume that 20 per cent. of the total pumpage cannot be accounted for, an assumption not unreasonable when viewed in the light of the experience of other cities using pumped supplies and with complete meterage, there should be deducted from the 56 gallons 20 gallons per capita per day, leaving 36 gallons to represent the domestic consumption and water used for public purposes. The amount of water used for public purposes in Cleveland is greater than in many of our large Eastern cities. No exact local data are obtainable, but judging from the conditions in other cities an allowance of about 10 gallons per capita would probably be sufficient. The total daily per capita consumption on these conditions would therefore be constituted as follows: Water for manufacturing and trades purposes................ 40 Water for domestic consumption............................ 26 Water for public uses....................................... 10 Under-registration of meters, leakage from mains, water used at plant, slip of pumps, and all other losses............... 20 Total.............................................. 96 In his report for 1904 Mr. Bemis reaches, on the following basis, the decision that about 30 per cent. of the pumpage for that year was wastage: Total metered water, 24,206,160 gallons per day or, 54.4 gallons per capita. Of this, 35.3 gallons per day are used by 654 large consumers, leaving 19.2 gallons per day per capita as the 141 Waste of Water and Its Reduction. amount used by the 23,658 small metered services. If the 30,000 remaining unmetered services would use water at the same rate, after metering, the balance of the consumption per day would be 5,760,000 gallons, making a total of 29,876,000 gallons for the entire city. This is 48.5 per cent. of the pumpage in 1904. He then estimates the public uses to be not over 6 per cent., and the loss by leakage, slip of pumps, etc., at 15 per cent., leaving 30 per cent. as the probable preventable waste, or, in tabular form: Gals. per cap. Water for consumers, domestic, and trades.............. 66.0 ". public uses................................. 8.0 " *" leakage, slip of pumps, etc................... 21.0 Probable total consumption............................. 95.0 Pumpage in 1904........................................ 136.0 Preventable waste from 1904 pumpage................... 41.0 The detailed effect of the introduction of meters is exhibited very clearly in the following diagram, which shows the average daily per capita consumption of water per inhabitant each month during 1900-1904 inclusive, and the number of meters in use at the end of each month for the same period. The cost of meter setting in Cleveland has been kept in great detail. For 1904 the cost of the /s-inch meters is given as follows: COST OF 8-INCH METERS IN CLEVELAND. COST PER METER Total Total Cost of Average NATURE OP Number. Cost of Each Cost For SETTING. Set. Setting. Meter. One Meter, Material. Labor. Completely Installed Basement..... 1,791 $0.57 $4.37 $4.94 $6.50 $11.44 Sewer pipe.... 2,584 3.75 5.64 9.39 6.50 15.89 Brick vaults.. 336 5.80 7.59 13.39 6.50 19.89 For each size set the average costs are said to have been as follows: 142 AVERAGE DAILY PER CAPITA CONSUMPTION OF WATER PER INHABITANT I Appendix D: Waste Reduction in American Cities. AVERAGE COST OF METERS IN CLEVELAND. Average Average Total SIZE. Nature of Setting. Number Cost of Cost of Average Set. Meters. Setting. Cost. 6...... Brick vault.8 $406.25 $43.50 $449.75 " "......... 108 226.46 38.25 264.71 4'... Box................. 8 226.46 33.00 259.46 Basement............. 15 226.46 21.19 247.65 Brick vault........... 148 91.26 30.54 121.80 3'... Box.................. 3 91.26 25.00 116.26 Basement............. 24 91.26 9.95 101.21 Brick vault........... 336 44.93 25.08 70.01 2... Box................. 8 44.93 18.00 62.93 Basement............. 83 44.93 11.51 56.44 Brick vault........... 385 37.37 18.23 56.60 1i.. Box.................. 2 37.37 18.00 55.37 Basement............. 102 37.37 12.03 49.40 Brick vault........... 738 27.01 16.25 43.26.... Box.................... Basement............. 204 27.01 8.52 35.53 Brick vault........... 1,569 15.74 16.61 32.35.... Sewer pipe........ 59 15.74 8.07 23.81 Basement............. 393 15.74 4.83 20.57 Brick vault........... 1,907 6.53 10.15 16.68..... Sewer pipe............. 12,529 6.53 8.23 14.76 Basement............. 11,225 6.53 3.96 10.49 The prices for the meters in the above table are those of recent purchases. The cost of setting is the average of the costs of 1903 and 1904. The total cost of all meters (29,854) was............ $311,157.50 The total cost of setting all meters was.............. 233,839.82 $544,997.32 The average cost per meter based on total number set of all sizes.................................... $18.25 The cost of maintenance and repairs of meters, including removals, replacements, damages due to frost, and to hot water, renewals of parts, etc., for labor and materials, was for 1903................................... $13,196.21 And for 1904, excluding the damages from frost............... 10,725.30 The damage from frost in 1904 cost to repair.................. 28,361.38 The damage from frost was confined almost entirely to the 58-inch meters set in sewer pipes, the protection afforded by the single cover not being sufficient to prevent the freezing of the meter and connections. An improved setting has been perfected by which the expense chargeable to freezing will be reduced to a minimum. Excluding the excessive damage done by frost 143 Waste of Water and Its Reduction. in 1904, the average cost of maintenance has been about 2Y2 per cent. of the cost of all the meters in use. In order to still further control wastage the Board of Public Service during 1904 metered practically all services larger than /8 inch, supplying public buildings, the Park Department, etc. Public and parochial schools are allowed the free use of an amount of water equal to 10 gallons per capita upon the attendance at the schools for the number of days the schools were in session; all water used in excess of that quantity is charged for at regular meter rates. Similarly, 40 gallons per capita per day are allowed free to certain classes of charities covered by the State law requiring free service, and 150 gallons per day per capita to the hospitals. Under a resolution passed September 2d, 1904, the Board of Public Service gives the Superintendent of the Water Works power at his discretion to meter, at the cost of the Water Department, all services, both residence and business, an ordinance having been passed in April of the same year in which it was provided that private residences should only be metered at the request of the consumer. This latter ordinance was passed following the scare arising from the, freezing of the meters, to restrict the Department from placing meters except when proof was to be had of the excessive waste, or where consumers desired meters. The effect of,the passage of the last mentioned ordinance was to stop, temporarily, the rapid introduction of meters and check the reduction of waste. During the latter part of the year the rate of introduction again increased, as will be seen from the diagram showing the per capita consumption and meters. The rate for metered water is uniform for all premises, as follows: ' 144 Appendix D: Waste Reduction in American Cities. METER FURNISHED AND SET BY CITY. Rate: 40 cents per 1,000 cubic feet. Minimum charges: 8-inch metered connection, $ 1.25 semi-annually where semi-annual assessment rate is $3.50 or less....( " 2.50 semi-annually in all other cases. 4 ( " ( 5.00 semi-annually. 1 " " " 6.00 " " 1 X2." 8.00 " 2 " " (i 10.00 " 3 ". " 15.00 " 4 " 20.00 " " 6 " " ( 30.00 " METER FURNISHED AND SET BY CONSUMER. Rate: 40 cents per 1,000 cubic feet. Minimum charges: Y4-inch meter...................................... $2.50 semi-annually All large meters................................... 4.00 " " All payments for water, either on metered or assessment basis, are made in advance upon an estimate of the Superintendent, based upon the probable use of water during the ensuing six months or less, as the case may be. If meters fail to register, or if they get out of order, the consumer is charged at the rate of average daily consumption shown by the meter when in order. All water passing through the meter is charged for, whether used or not. Upon a written request to the Superintendent by any consumer, accompanied by an inspection fee of $1.00 for meters 1 -inch and smaller, $3.00 for meters from over 1 inch and under 6 inches, or $10.00 for 6-inch meters, the Department will take the meter to its office and test it. If the meter is found to overregister the inspection fee is returned to the consumer and a reduction made in the current bill to correspond with the facts. If the meter does not over-register the fee is retained for the expense of testing. All meters are sealed and if there is good evidence that a meter has been tampered with the water is turned off and not turned on again until the payment of $1.00 and an advance deposit of twice the amount otherwise required has been made. 145 Waste of Water and Its Reduction. In such cases the owner or consumer also has to pay for the estimated quantity of water not registered because of the tampering with the meter. DETROIT, MICH. NTIL about 1868, the per capita consumption of water in Detroit had been less than 60 gallons per day. After that year the rate of consumption increased almost constantly, until twenty years later it had reached an average of 210 gallons per capita per day, with a maximum on certain days of 250 gallons. During this period the population had increased 158 per cent., while the amount of water pumped had increased 755 per cent. and the amount pumped per capita 234 per cent., an unreasonable increase, since the character of the city had not changed sufficiently to account for the excess in a legitimate manner. In 1899 the city was confronted with the alternatives of either increasing the capacity of the plant or reducing the waste of water, and after careful consideration the Board of Water Commissioners directed Mr. L. N. Case, then Secretary and afterward General Manager, to co-operate with the Superintendent in devising means of preventing the rapidly increasing waste. In April, 1899, therefore, eight inspectors were appointed for house-to-house inspection, to locate leaks and leaky fixtures. Previous to starting the work of inspection a large number of circulars, urging the citizens to co-operate with the Water Board in suppressing the needless waste, had been printed, and were distributed to every dwelling, workshop, store, office, factory and manufacturing place in the city. The result of this appeal was quickly felt in the pumping station, the amount of water pumped per day beginning to decrease almost immediately. When the house-to-house inspections were inaugurated, preparations were made for the placing of meters on all properties having partial or indefinite supply. In order that proper control could be had over the different services, a systematic 146 Appendix D: Waste Reduction in American Cities. examination was made of all shut-off boxes and those found out of order were repaired or replaced. This inspection bureau was maintained in an active, aggressive condition for several years; but after 1897 the duties of the inspectors had increased to such an extent that it became impossible for them to give as much attention to the finding of leaks and detection of waste as they had done in previous years. The work of the inspectors was of great value not only in finding and stopping waste and leaks, but also in locating many fixtures of which the Department had no record and for which no revenue had been derived. When a leak or an excessive wastage was found a notice would be served upon the householder stating a date upon which the second inspection would be made and if the leak had not been stopped on the second visit the water would be shut off until the plumbing was put in good order. In cases where the poverty of the householder could be urged as an excuse for the delay, extensions would occasionally be granted and a third visit made at a later date. Meters were first applied to livery stables, laundries, breweries, saloons and manufacturing establishments, and later to the principal stores and to dwellings where excessive waste was detected. Within two years from the time metering was first commenced many meters were placed on premises at the request of the householders; and by 1897, upwards of one thousand were in use on the services of dwellings. In that year, owing to the discrepancies between the meter rates and assessment rates for domestic uses many of the houses were restored to assessment rates, the meters being allowed to remain on the services as checks on the wastage. Since 1897 these meters have been gradually removed and placed on business properties, so that in 1905 but 171 meters remained in place on the services of premises which had been restored to assessment rates. The metering of domestic services practically stopped in 1897. In 1896 investigations were made, with Deacon meters, to locate districts of excessive wastage. These investigations brought out, prominently, that most of the water wasted in the districts examined was chargeable to a very few careless and ex147 Waste of Water and Its Reduction. travagant consumers, the majority of the consumers keeping their plumbing in order and being reasonably careful not to unnecessarily waste the water. As in all other cities where such investigations have been made the water-closet and the leaky faucet were found to be among the principal sources of waste. The placing of meters has had the effect of reducing waste in the houses and also of aiding in locating leaks in service pipes, between the meters and the house-piping, many of these having been underground and invisible, although, in some cases, of considerable magnitude. In 1894 the meter system was extended to public buildings, hospitals and charitable institutions, resulting in the checking of enormous wastage. Since 1897 the practice of allowing domestic consumers to pay according to assessment rates instead of meter rates has resulted in the gradual removal of the domestic meters and the placing of the meters thus removed upon the services supplying places of business. In many premises which had been placed on the assessment roll the meter showed that the occupants had again become recklessly wasteful, the consumption jumping up from ten or twelve thousand gallons per quarter to nearly three hundred thousand gallons in the same length of time. Such places would be restored to meter rates until the consumption had again been reduced to a legitimate amount. The effect of the meters in restraining waste is well shown by the following comparison of the per capita consumption in 35 residences, for a year, paying by assessment rates and the same houses, for the same length of time, paying by meter rates. Paying by assessment rates,-106 gallons per diem per capita. 4" " meter - 45 " " " " " In the houses when paying by assessment rates the meters were still in place and the consumption measured thereby and their presence undoubtedly exerted a restraining influence on wasteful tendencies. In 1901 it became necessary to transfer 176 consumers back 148 Appendix D: Waste Reduction in American Cities. to the meter rates on account of their large consumption. In the case of one private family the meter indicated a consumption of 435,000 gallons in three months, while after transference back to meter rates the consumption decreased to 9,000 gallons for the same length of time. In 1902 sixty premises were placed back on meter rates and in 1903, 41 premises. Since that year there is no record in the published reports of the re-establishment of meter rates in the case of domestic consumers transferred to assessment rates. Table 28 (see next page) gives the quantity of water pumped each year, since 1860; the per capita consumption, based on the U. S. Census returns for 1860, 1870, 1880, and 1890, and the State Census of 1894;/the number of meters in place each year since their introduction in 1888, and the percentage of the total pumpage measured through meters. In 1888 the average consumption for the year was 210 gallons per capita per day. In 1897 by persistent and careful inspection, stringent enforcement of regulations and the placing of meters on wasteful consumers the consumption had been reduced to 136 gallons per capita per day, with only 5,400 meters in service and about 20 per cent. of the consumption metered. During the latter part of that year began the transference back to assessment rates of domestic consumers who had previously paid by meter rates; and thi( practice has continued up to the present time, so that in 1905 come paratively few domestic consumers were paying by meter rates. Thus, while the total number of meters in use has steadily increased each year the increase since 1897 has been on business and manufacturing consumers while the number of metered domestic consumers has decreased. Since 1897 the daily per capita consumption has steadily increased until in 1905 it had reached an average figure as high as prevailed in 1899, the year when waste restriction was undertaken, in spite of the fact that there were more meters in use than in 1897, which was the year of lowest relative consumption since 1881. The explanation of this increase in consumption may possibly be found partly in the relaxation of the work of house-to-house inspection and perhaps more largely in the taking out of the meters formerly placed on wasteful domestic consumers. 149 Waste of Water and Its Reduction. TABLE NO. 28. PUMPAGE, CONSUMPTION AND METERS, DETROIT. POPULATION USED. C. 45,619 49,000 52,400 55,800 59,200 62,600 66,000 69,400 72,800 76,200 C. 79,577 83,200 86,900 90,600 94,300 98,000 101,700 105,400 109,100 112,800 C. 116,340 125,000 134,000 143,000 152,000 161,000 170,000 179,000 188,000 197,000 C. 205,876 213,000 221,000 229,000 237,000 245,000 253,000 261,000 269,000 277,000 C. 285,704 296,000 307,000 S. 317,600 327,000 WATER PUMPI IN GALLONS YEARS. Total Quantity. 1860................. 870,036,451 1861................ 895,129,423 1862................ 994,945,329 1863................ 1,035,798,043 1864.............. 1,019,390,256 1865............ 1,040,514,887 1866................ 1,196,317,922 1867............ 1,425,535,230 1868................ 1,666,545,125 1869............ 1,946,810,325 1870................ 1,866,060,068 1871............. 2,300,150,605 1872............ 2,782,292,578 1873................3,198,393,948 1874............... 3,289,872,635 1875................. 4,207,454,260 1876................ 4,065,134,470 1877................. 4,213,239,790 1878................ 4,345,743,330 1879........... 5,129,599,110 1880................ 5,552,965,310 1881................. 6,543,127,968 1882................ 6,284,000,742 1883............ 7,379,327,788 1884.............. 8,510,614,140 1885........... 9,970,829,580 1886................ 10,576,571,254 1887............... 13,168,859,808 1888........... 14,380,166,670 1889................. 12,875,334,453 1890................ 12,120,944,532 1891................ 12,057,261,236 1892................ 12,276,612,482 1893............ 13,877,977,208 1894............ 13,649,779,605 1895............ 14,698,451,954 1896............ 13,254,369,371 1897................ 12,928,821,326 1898......... 14,278,682,058 1899................. 16,348,062,701 1900........... 17,077,100,477 July, 1901, to July, 1902 18,333,104,706 July, 1902, to July, 1903 18,757,682,360 July, 1903, to July, 1904 21,323,839,222t July, 1904, to July, 1905 21,566,461,863t 52 50 52 51 47 46 50 54 63 70 64 76 88 97 96 118 110 110 109 125 131 144 129 141 154 170 170 202 210 180 161 155 152 166 158 165 144 136 146 162 164 170 168 183 181 SD, 3. No. ofPercentMe- age of ters Supply Per in Passe Capita, Place. through Daily.* Meters. 194 866 1,239 3186 3,775 4,585 5,392 5,470 5,555 5,635 5,738 5,847 5,957 6,076 *.r........ * @ @........... *.. *..... *. - * -~ 6,076 0.6 1. o5. 10. 13. 14. 17. 22. 23. 25. 26. 27. *ooo * - o.... 1....... 1.... Population: C=U. S. Census. S=State Census. *Based on population given. t411,115,012 gallons delivered outside Detroit; pumped. 150 deducted from total Appendix D: Waste Reduction in American Cities. THE FOLLOWING TABLE SHOWS THAT AS THE NUMBER OF METERS REMOVED FROM DOMESTIC CONSUMERS AND PLACED ON BUSINESS SERVICES INCREASED, THE PER CAPITA CONSUMPTION OF WATER INCREASED GREATLY COINCIDENTLY THEREWITH. Number of Total No. Number of Consumers of Meters Daily Total No. Meters From Restored to Removed Average YEAR. of Meters Which Assessment from Do- Consumption in Place. Revenue Rates, Meters mestic Con- Per Capita, was being still sumers, to Gallons. Derived, in Place. Year Indicated. 1894........ 3,186 3,188............ 158 1895........ 3,775 3,775............ 165 1896... 4,585 4,585............ 144 1897........ 5,392 5,392............ 136 1898........ 5,470 4,325 1,045...... 146 1899........ 5,555 4,442 1,113...... 162 1900-01*.... 5,635 4,828 807 306 164 1901-02..... 5,738 5,081 656 457 170 1902-03..... 5,847 5,337 510 603 168 1903-04..... 5,957 5,774 373 740 183 1904-05..... 6,076 5,905 171 942 181 *Fiscal year changed 31st, 1900. to end June 30th, 1901, instead of December In succeeding years, fiscal years end June 30th. Thus it will be seen that coincident with the allowing of domestic consumers to return to assessment rates the per capita consumption began to increase and continued to increase coincidently with the removal of the meters from such consumers until it had, in five years, reached approximately the figures obtaining fifteen years earlier. The temptation is very strong to accept this explanation. It is a fact, however, that the rate of consumption in Detroit has been increasing, from some legitimate, undetermined cause, as the increase of metered consumption in 1899 over 1898 was greater than the increase in the unmetered consumption, while the capacity of the meters in use in 1899 was practically the same as in 1898. 151 Waste of Water and Its Reduction. Total metered water in 1899................. 3,425,487,986 gals. " " " " 1898................. 2,792,842,000 " Increased use of metered water.............. 22.6 per cent. Total unmetered water in 1899............... 12,867,900,465 gals. " " " " 1898............... 11,470,788,658 " Increased use of unmetered water............ 12.2 per cent. In 1899 21 per cent. of the total supply was metered. In 1898 19 " " " " " On examining the reports of the inspection work it is apparent that the amounts charged to inspections since 1897 have been nearly double, per capita, the amounts charged prior to that time, while the reports of work done by the inspectors indicate that after 1897 the number of inspections made to discover leaks has been, on the average, only about one-third as many per year as during the years prior to 1897, and the number of leaks reported proportionately smaller. On the other hand, the inspectors have had numerous additional duties assigned to them as the result of the change in the policy of the department. Up to and including 1897 the cost of the meters and their instaflation amounted to $158,925, or an average of about $29 per meter. Since that time about 950 meters have been taken off of domestic consumers and reset on business properties and nearly 750 new meters have been furnished and set. It is impossible, therefore, to arrive at the cost of meters since 1897. The total amount charged to the meter account for meters and setting, from 1889 to 1905 inclusive, has been $234,923.78, during which time there were set about 6,000 meters, many of these having been originally placed on the premises of domestic consumers and in later years removed and reset on business properties. The net results of the efforts at waste reduction in Detroit may therefore be summed up as follows: For 9 years, from 1889 to 1897 inclusive, under the general management of Mr. L. N. Case, the per capita consumption was gradually reduced from 210 gallons to 136 gallons per day, a reduction of about 74 gallons per capita or about 35 per cent, while the population of the city had increased about 33 per cent during the same length of time. 152 Appendix D: Waste Reduction in American Cities. This result was accomplished by complete inspections of all premises and the placing of meters on wasteful consumers. So long as this policy was continued the wastage was controlled to such an extent as to keep the per capita consumption below 165 gallons per day. The average cost for this work, for the meters, their installation, and the inspection bureau was about $21,000 per year. The estimated value of the water saved during that time, assuming the per capita consumption to have remained at the figure reached in 1889, was about five times the cost of the Meter and Inspection Bureaus. The good results thus obtained have not been permanent, as, although practically the same amount has been spent annually during the seven years from 1898 to 1905 inclusive as during the earlier period, the per capita consumption has again rapidly mounted to almost as high figures as prevailed before waste reduction methods were put in force, although more meters are in service than during the earlier period. Whether this increase has been entirely chargeable to the smaller number of inspections made and to the restoring to assessment rates of domestic consumers who were previously placed on the meter rolls is not entirely proven. That the circumstances are significant there can be no doubt and the only way by which the connection could be conclusively proven would be to adopt again the former policy of metering wasteful domestic consumers, resume, persistently, the house-to-house inspections and observe the results. If the removal of about 1,000 domestic meters and the relaxation of inspections has caused this apparent relapse to the oldtime wastefulness the cost of placing these meters back, with $10,000 per year additional for the cost of extra inspections, would be well repaid. The situation, however, is a difficult one, for the minimum meter rates for domestic consumers is $7 per year while the majority of houses, at assessment rates, would pay a smaller amount. There is, therefore, little incentive on the part of domestic consumers to have meters placed on their services, and although the 153 Waste of Water and Its Reduction. authorities can place meters where wastefulness prevails they cannot compel the domestic consumer to allow the meter to remain if the meter readings do not show wastefulness or extravagant use. A wasteful consumer on whose service a meter is placed can, therefore, by preventing waste for a short while have the meter removed and again revert to his former careless wastefulness when restored to assessment rates. It is impossible, from the Annual Reports of the Department, to determine whether or not the increase in consumption for business and manufacturing purposes will account for a significant proportion of the total increase, but with only about 26 or 27 per cent of the total supply metered it seems unlikely that it can all be accounted for in this manner. FALL RIVER, MASS. N Fall River waste reduction began more than thirty years ago. In 1874 when there were but 6,000 consumers of Wautuppa Lake water the per capita consumption among the consumers was 84.53 gallons per day, only about 8 per cent. of the services being metered. Each year thereafter more meters were added until, in 1879, with about 55 per cent of the services metered the consumption had dropped to 35.6 gallons per capita, a reduction of 58 per cent. In 1882 with 63 per cent of the services metered it climbed up again to 45.9 gallons per capita, falling gradually to 26.89 gallons in 1887 with 70 per cent of the services metered. From 1887 to 1902 it has remained at between 30 and 40 gallons per day. The following table contains the number of consumers, average daily consumption of water, average daily consumption per consumer per day, number of services, number of meters in use, and percentage of meters to services each year from 1874 to 1902. 154 Appendix D: Waste Reduction in American Cities. TABLE NO. 29. SUPPLY FROM WAUTUPPA LAKE; PUMPED. YEAR. 1874.... 1875.... 1876.... 1877.... 1878.... 1879.... 1880.... 1881.... 1882.... 1883.... 1884.... 1885.... 1886.... 1887.... 1888.... 1889.... 1890.... 1891.... 1892.... 1893.... 1894.... 1895.... 1896.... 1897.... 1898.... 1899.... 1900.... 1901.... 1902.... Average Daily Consumption. 507,168 810,980 1,057,704 1,173,601 1,204,217 1,263,925 1,353,641 1,488,247 1,830,801 1,640,481 1,425,861 1,488,137 1,603,482 1,590,960 1,768,524 1,877,937 2,136,182 2,355,700 2,285,948 2,333,888 2,438,231 3,166,509 3,547,287 3,669,640 3,136,049 3,580,895 3,804,867 3,618,739 4,365,065 Number of Consumers. 6,000 11,450 22,000 28,000 33,000 35,500 36,440 38,120 39,891 43,091 43,733 48,200 50,420 59,155 60,524 64,000 69,000 71,000 76,000 81,000 85,576 86,076 94,000 97,500 94,267 98,931 104,523 106,631 107,653 Average Daily Consumption per Consumer. 84.53 70.38 48.08 41.91 36.49 35.60 37.15 39.04 45.90 38.07 32.60 30.87 31.80 26.89 29.22 29.34 30.96 33.18 30.08 28.81 28.49 36.79 37.74 37.64 33.26 36.20 36.40 33.94 40.54 Number of Services. 672 1,147 1,660 2,066 2,324 2,497 2,685 2,906 3,120 3,370 3,611 3,818 3,986 4,197 4,412 4,698 4,980 5,247 5,526 5,793 6,138 6,372 6,704 6,422 6,576 6,783 6,943 7,075 7,282 Number of Meters. 53 193 585 881 1,165 1,372 1,583 1,780 1,966 2,187 2,421 2,569 2,725 2,941 3,138 3,428 3,717 3,975 4,252 4,529 4,887 5,212 5,607 5,954 6,128 6,363 6,544 6,755 6,973 Percentage of Meters to Services. 8 17 35 43 50 55 59 61 63 65 67 68 69 70 71 73 75 76 77 78 80 82 83 93 93 93 94 95 95 1902 number of consumers per service = 14.8. In connection with the introduction of meters the Wautuppa Water Board has always exercised a close watch over the consumers, both metered and unmetered, to keep wastage and surreptitious usage down to the lowest practicable limit. In recent years, the superintendent, Mr. Patrick Kieran, has accounted for the various uses of water in great detail, and his reports are models worthy of adoption by other Water Works,superintendents. It is, unfortunately, impossible to dig out, from about 95 per cent of the Annual Reports of the Water Boards of cities in the United States, useful, reliable or intelligible data; 155 Waste of Water and Its Reduction. and it is a positive pleasure and relief to occasionally come upon such valuable, detailed information, so well arranged, and bearing such a genuine stamp of honesty, intelligence and purpose as is to be found in the Annual Reports of the Wautuppa Water Board. This is one city where an earnest effort is being made to ascertain how much water is supplied, how much reaches the consumers, how much is used for various public and non-revenue productive purposes, and how much is lost or can not be accounted for. The statements are not only interesting, but of great value. As to the general results that have been obtained since water has been largely sold to consumers by measure, it may be stated that for the past 28 years the per capita consumption has been continuously kept at a figure less than half the rate that prevailed prior to the time the change was made in the system of selling water. The lowest point was reached when about 70 per cent of the services were metered. Since that time there has been a slight upward tendency due to increased use of water with increased population and the increased necessities for the use of water characteristic of all growing cities. This small increase is rational and natural and shows that the use of meters does not check the use of water, but does check the useless wastage due to carelessness in not keeping fixtures in proper order. No better data can be found bearing upon the merits of the method of charging by measure for water than in Fall River. Exclusive of the manufactories, mills, etc., the majority of users are renters paying for their water in the rent, the direct payment for the water falling on the owners of the properties supplied; and yet, Without any incentive to be over-careful, these renters, when their plumbing is kept in repair by the owners, do not waste water and have not done so for nearly thirty years. This is the experience of every city that has sold water by measure, in every country. From the Reports from 1899 to 1902, inclusive, the data regarding the unaccounted-for water have been arranged in the following tables: 156 Appendix D: Waste Reduction in American Cities. USE OF WATER FOR DIFFERENT PURPOSES. AVERAGE PERCENT- ACCOUNTED FOR WATER. DAILY AGE OF _ CONSUMP- TOTAL TION, GALS. PUMPAGE. Percentage Percentage Not EAR. Popu- N f Metered. Metered. Y lation. Consumers. -.- ----------- I -... co o.. 4____4__Z__Og &l r v3 p4 Q s ~ 1899.... 102,281 98,93135.0136.20 76.4723.53 74.0 8.8 82.8 5.4 11.8 17.2 1900.... 107,6232 104,5235.3536.4078.2921.71 69.3 9.1 78.4 6.0 15.6 21.6 1901.... 107,831 106,63133.5633.9480.6919.31 68.6 4.7 73.3 6.4 20.3 26.7 1902.... 108,728 107,65340.1540.5478.5321.47 68.3 5.2 73.5 5.8 20.7 26.5 PER CAPITA AMOUNTS TAKEN BY CONSUMERS, METERED AND UNMETERED; PER CAPITA AMOUNTS PER CONSUMER, USED FOR PUBLIC PURPOSES, AND AMOUNTS NOT ACCOUNTED FOR. X X, ' GALLONS USED PER DAY, GALLONS USED PER ER CONS USED PER Gallons D PER CONSUMER fOR Total 0,v DAY PER CONSUMER. PUBLIC PURPOSES. per Day Supply ____ per Conper Con-, o sumer, YEAR. smer, Me- Not Me- Not not AcGals. "< c: tered. Me- Total. tered. Me- Total. counted per Dayllons tered. tered. For. Gallons 1899 36.20 27.7 20.5 1.5 22.0 2.4 3.3 5.7 8.5 1900 36.40 28.5 19.7 1.7 21.4 2.6 4.5 7.1 7.9 1901 33.94 27.3 18.7 1.7 20.4 1.3 5.6 6.9 6.6 1902 40.54 31.8 21.7 1.8 23.5 1.6 6.7 8.3 8.7 During these four years, when careful inspections were made to discover leaks and wastage, when all water used for public purposes was either metered, measured by the capacity of the tanks used for flushing sewers, watering streets, etc., or estimated by partial metering of representative portions of the water used, and the unmetered consumers' water estimated by careful comparisons with metered consumers, it was still impossible to account for about 21.5 per cent of the total pumpage. It is probable that in no American city has there been a more earnest endeavor to limit the amount of unaccounted-for water than in Fall River, where practically all the water used by consumers is metered; and this statement, showing that at least one-fifth of the 157 Waste of Water and Its Reduction. water pumped is not accounted for by the meters, after allowing a correction of 5 per cent to the metered water for possible underregistration of the meters, may be taken as a fair index of the loss that may be expected, even with the greatest care and watchfulness to prevent wastage. Although Fall River is a city the interests of which are largely devoted to manufacturing, the amount of water used by these factories is relatively small as compared with many other cities, Harrisburg, Pa., for instance, where the per capita daily amount of water sold by meter for manufacturing and railroad purposes alone is twice as great as the total per capita daily amount of water sold by meter and by flat rates combined, to all consumers, domestic and manufacturing, in Fall River. This, again, emphasizes the impropriety of comparing the per capita water consumption of different cities for the purpose of showing that because one city has a low per capita consumption of water, due to the use of meters, therefore all other cities having high rates of consumption should by using meters, cut down their consumption to equal figures. It will be noticed that after a sufficient proportion of the water was sold by measure to suppress excessive wastage the effect of the sale of water by measure has been to suppress permanently undue waste, but not use, as evidenced by the relatively uniform and comparatively slow increase in per capita consumption, and the absence of erratic or excessive fluctuations in rate from year to year. The cost of the meters and cost of setting and repairs falls on the consumers; during the years 1899 to 1902 inclusive the total annual expense for meters and their maintenance averaged about one-half of one per cent of the value of the meters in use. HARRISBURG, PENNSYLVANIA. URING 1888 the Board of Water Commissioners, facing a rapidly increasing consumption of water, instituted a rigid house-to-house inspection for the detection of leaks, and also adopted the policy of placing additional meters on large consumers, with the object of reducing the waste which was believed to be excessive and increasing rapidly. The placing of meters,wl58 Appendix D: Waste Reduction in American Cities. began a year or two before this, and the number installed had not been sufficient to affect the total supply, only 92 meters being in use in 1882, of which 81 were owned by the consumers and 11 by the city. Since that time there have been added from 300 to 400 meters each year until in 1904 there were 5,932 in use, 71 of which werq owned by the city. During that year 8,359 buildings out of a total of 13,177 were supplied through meters with the city water. Metering in Harrisburg is voluntary, no legislation having been enacted making the use of meters compulsory, the consumers, except in the case of large manufacturing establishments, purchasing the meters from a stock of various makes kept on hand by the Water Department. Several well-known reliable makes of meters were adopted for introduction, and the manufacturers furnish such numbers as are desired by the Board. The meters thus received are placed in a fireproof vault at the pumping station and remain the property of the manufacturers, to be paid for when sold to a consumer. The consumer is charged about 15 per cent more than the cost of the meter to cover the labor of superintending the setting. The meter is not delivered to the consumer, but, after having paid for it at the City Treasurer's office, he leaves the receipted permit at the office of the Water Board stating the hour he wishes the meter set. At the appointed hour the proper officer of the Water Department takes the meter to the place required, where it is set by a licensed plumber under the supervision of the officer. The purchaser selects voluntarily, and without suggestion from the Department, the kind of meter he wishes. The officials of the Water Board neither receive nor pay out any moneys in connection with the transaction. Meters in use which are suspected of improper registration are removed by the Water Department, tested, and if found correct, replaced without cost to the consumer. If repairs are necessary, however, they are made by the Department, if possible, and if not, they are sent back to the factory. The cost of all repair work is charged to the consumer. When a meter is removed for any purpose a pipe connection is inserted so that the supply is continued while the meter is out. If not promptly replaced the water used in the interim is charged for at the same rate as the previous quarter's consumption. The domestic meters are read quarterly; all 159 Waste of Water and Its Reduction. others are read monthly. Bills for the water consumed are delivered by the inspector in person, quarterly and monthly, for the two classes of consumers. On January 1, 1905, 64 per cent. of the residences in Harrisburg were taking water through meters. Practically all the manufacturing consumption is metered. The unmetered manufacturing and elevator consumption and water used for public purposes was estimated to be about 10 per cent of the metered manufacturing consumption. The population, average daily consumption of water, consumption per capita for domestic and manufacturing purposes and number of meters in use from 1885 to 1905 inclusive is given in Table 30. These records are among the most interesting and instructive to be found in published reports. It will be noticed that the per capita consumption is high for a city of this size, and yet, when it is separated into its component parts it is reasonable and rational. There are probably few cities in the United States which show so large a proportion of the total consumption devoted to manufacturing uses. From 1888 to 1899 the total per capita consumption remained practically stationary, varying only 5 or 6 gallons either way from 122 gallons per day. The effect of the progressive introduction of meters is shown in the gradual reduction of the per capita domestic consumption from 89 to 55 gallons per day, the curtailment of the waste represented by this difference being sufficient to supply the gradually increasing demands of the manufacturing consumption, which had grown from 33 to 65 gallons per day in the same length of time. Since 1899 the manufacturing consumption, which is all metered, has continued to grow as the city's resources have developed, while the domestic consumption has not varied more than would naturally be expected, the increase in the last two years having been due to the very cold winter weather for three months in each of these years, as shown by the pumping records. The per capita manufacturing consumption alone in Harrisburg, in 1904, was, therefore, five times the average total per capita consumption in Berlin for all purposes, domestic, commercial, manufacturing and public. The average cost of furnishing and setting the meters has been 160 Appendix D: Waste Reduction in American Cities. TABLE N0. 30. CONSUMPTION AND METER DATA, HARRISBURG. AVERAGE DAILY Average NUMBER OF CONSUMPTION PER Popu- Da METERS IN USE. CAPITA, IN GALLONS. YEAR. lation. Consumption, Gallons Do- ManuA. B. Total. mestic. facturing. 1885....... 36,000 4,462,000....... 112........ 1886........ 37,000 5,130,000....... 126........ 1887....... 38,000 6,400,000....... 151........ 1888....... 39,000 6,336,000 81 11 153........ 1889....... 40,000 5,557,000 228 8 122........ 1890....... 41,000 5,859,000 504 11 128........ 1891....... 41,000 5,762,000 831 10 122 89 33 1892....... 42,000 5,901,000 1,165 14 122 79 43 1893....... 43,000 6,054,000 1,523 16 121 79 42 1894....... 44,000 5,528,000 1,932 19 108 71 37 1895....... 45,000 6,308,000 2,401 19 119 75 44 1896........ 46,000 5,853,000 2,832 22 107 65 42 1897....... 47,000 6,012,000 3,274 31 107 60 47 1898....... 48,000 6,489,000 3,622 41 112 63 49 1899...... 49,000 7,323,000 3,911 46 122 57 65 1900....... 50,000 8,203,000 4,208 52 133 60 73 1901....... 52,000 8,767,000 4,548 55 135 55 80 1902....... 54,000 9,064,000 5,046 59 134 55 79 1903....... 57,000 8,052,000 5,502 63 143 65 78 1904....... 60,000 8,787,000 5,861 71 146 65 81 Population based on census returns, 1880, 1890 and 1900. Prior to 1903 no corrections made in pumpage records for slip. Slip of old pumps found to be 25 per cent. in 1903. Total per capita consumptions given are based on allowances from 20 per cent. to 12 per cent. for years prior to 1903. Manufacturing consumption is practically all metered and domestic consumptions given, which include the unaccounted for water, are differences between total and manufacturing uses. Manufacturing uses given include water used for public purposes. A. Meters on domestic services, and owned by consumers. B. Meters on manufacturing consumers, and owned by City. 161 Waste of Water and Its Reduction. about $13.50 per meter, and the cost of repairs and maintenance about 1 per cent per annum on the value of the meters in use. SUMMARY.-The gradual introduction of meters in Harrisburg, Pa., until practically all the manufacturing and about 65 per cent of the domestic consumers are metered, has had the effect of curtailing domestic wastage to the extent of about 30 gallons per capita per diem. This result has been accomplished largely by the voluntary metering of services by consumers under meter rates by which the domestic consumers could pay by meter somewhat smaller annual charges than under the prevailing fixture rates. In conjunction with the placing of meters considerable inspection work is done to keep down leakage and losses, and all leaky street mains are repaired as soon as discovered. The high per capita rate of consumption is reasonable when it is understood that 55 per cent of the total supply is used for manufacturing and public purposes, the public uses being comparatively small, certainly not over 3 or 4 per cent of the total. While the total per capita consumptions given are not claimed to be correct, and cannot be when, in a pumped supply the slippage of the pumps is not checked up by measurement, it is believed that for purposes of comparison the relative ratios of domestic, manufacturing and total consumptions are sufficiently exact. Although at the present time the per capita consumption is as high as in 1887, when waste reduction methods were first introduced, this is accounted for by the increasing quantities sold yearly to manufacturers and railroads, and is not due to waste, as was formerly the case. This is a case in which, without understanding the conditions, the high rate of consumption would be charged to enormous wastage. There is still room for improvement, and if the checking of waste on the present 35 per cent of unmetered domestic consumers produces an effect proportionate to that from the metering of the first 65 per cent, the reduction yet to be looked for would be about 10 gallons per capita per day on the domestic consumption, thus bringing it down to 45 or 50 gallons. 162 Appendix D: Waste Reduction in American Cities. IHARTFORD, CONNECTICUT. IN the early months of 1900 the waste of water in Hartford had reached such proportions that the Board of Water Commissioners found it necessary to appoint a force of ten inspectors and start them out on a house-to-house inspection, one inspector to each ward in the city, to examine all water pipes and fixtures and report such waste and leaks as were found. The result of the inspection was the discovery of a large number of closets and faucets wasting great quantities of water, as well as of broken service pipes, ruptured by freezing, and others kept running to protect them from frost; the latter was the greater difficulty to contend with, as some of the landlords insisted upon the waste of water by tenants as a protection for the plumbing during freezing weather. The Board had also under consideration the propriety of installing a meter system, as it was estimated that fully onethird of the supply was wasted and some means of limiting the waste was imperative under existing conditions. After proper consideration it was decided to adopt universal metering, and accordingly the placing of meters was begun about the 1st of July. Up to the 1st of March, 1900, only 550 meters were in use in the city. During the fiscal year ending March 1, 1901, 2,233 additional meters were set, making the aggregate at the end of the year 2,783, with a total of 9,656 services in use. During the two succeeding years about 6,373 additional meters were set, making a total of 9,156 meters and 10,083 services. The placing of these meters, following the house-to-house inspection made in 1900, resulted in the finding and stopping of many leaks underground which would not otherwise have been detected. It was shown from the records of the Venturi meter, through which the supply to the city was measured, that in 1899 the night consumption between 1 A. M. and 4 A. M., February 23d to March 1st, averaged at the rate of 87 gallons per capita per day, while, from February 12th to 18th, 1902, between the same hours, the consumption was at the rate of about 55 gallons per capita per day. The difference, 32 gallons per capita, being a fairly good indication of the amount of wastage curtailed. This is also indicated by the difference, 38 gallons, in the average daily per capita 168 Waste of Water and Its Reduction. consumption for the years before general metering was undertaken and the consumption for 1902-03. Much of the leakage was from the service pipes between the mains and the houses, the investigations disclosing corrosion and pitting by electrolysis. The cast-iron mains were rarely found to be attacked. Out of 58 examinations made in 1903-04 the damage found was, with but three exceptions, confined to the galvanized iron service pipes, in connection with which it developed that many of the seams were imperfectly welded, the defects being concealed by the galvanizing, and when the zinc was removed by corrosion the seam opened and a leak resulted. The service pipes, with their zinc coatings were, at one end, mechanically connected with the main pipes to which both wrought iron and zinc are electrically positive. This combination forms a galvanic couple having a tendency to oppose the passage of current from the cast-iron pipe to the soil. The cast-iron pipes are coated and therefore, to a certain extent, insulated, while the service pipes are of wrought iron galvanized with zinc, a positive metal to come in contact with the soil. During 1902-03 three and one-half miles of street mains, from 4 inches to 12 inches diameter, were tested for leakage by shutting off the pressure at night on a section of main and bye-passing one of the closed main-valves through a one-inch meter by means of: hose. The shut-off boxes on the service pipes being closed temn porarily, the flow of water found, if any, represented the leakage from that section of the mains and from the service pipes between the main and the shut-off boxes. Also, in addition, the service pipes were tested between the shut-off boxes and the mains on about 134 miles of streets. The aggregate leakage found and cured amounted to about 150,000 gallons per day, or, figuring it upon the 312 miles of mains examined, about 40,000 gallons per day per mile of main, the indications being that most of the leakage was from the service pipes. In 1903-04 twelve miles of mains were examined for leaks, 50 leaks being found, 34 being in service pipes and 3 in the mains. Considerable leakage came from defective hydrants. In 1904-5 the mains and services were tested on 44Y2 miles of streets, in addition to going over a second time the 15y2 miles upon which the tests were made in the two previous years. One 164 Appendix D: Waste Reduction in American Cities. hundred and nineteen leaks were found, of which 20 were in service pipes on premises, 43 in service pipes between the main and the shut-off cocks, 54 were from defective hydrants, one was in the mains and one was from a defective gate valve. The amount of leakage cured was not determined. The following table gives the average consumption, number of services in use, number of meters placed, daily average consumption per service, and estimated per capita consumption from 1898 to 1905: TABLE NO. 31 CONSUMPTION AND METER DATA, HARTFORD. AVERAGE DAILY CONAverage Daily SUMPTION, IN GALLONS. YEAR. Water Con- No. of No. of sumption, Services. Meters. Gallons. Per Per Service. Capita.** 1897-98......... 9,046,000 8,672 497 1,105 110 1898-99......... 9,345,000 9,021 530 1,035 110 1899-00......... 9-10,000,000 9,351 550..... 1900-01......... No data 9,656 2,783 No data No data 1901-02......... u 9,914 6,993... 1902-03......... 6,981,000 10,083 9,156 692 72 1903-04.......... 6,988,000 10,239 9,604 635 70 1904-05......... 6,348,000 10,006* 9,860 635 62 * Several abandoned services shut off and discontinued. ** Deduced from estimated population. The question of establishing a flat meter rate was investigated by a commission in 1901. It was found that under the conditions prevailing in Hartford a flat rate was not equitable. There were 12 manufacturing consumers who used very large quantities of water, and who had been paying for this at a special rate of Y4 cent per 100 gallons. To put the whole City on this rate would not produce sufficient revenue to meet the expenses of the Department. On the other hand, the best flat rate that would produce the necessary revenue was 1134 cents per 100 cubic feet, or 1.57 cents per 100 gallons, a price too high for these large manufacturing consumers to accept without serious loss. It was therefore felt necessary to establish two rates, as follows: A.-12 cents per 100 cubic feet for consumers using less than 3,000 cubic feet, or 22,500 gallons per day. 165 Waste of Water and Its Reduction. B.-6 cents per 100 cubic feet for consumers using more than 3,000 cubic feet per day. Under these rates the small consumers only pay about 1-30th of a cent per 100 gallons more than if the lowest possible flat rate had been adopted, while the large users were given a rate but very little higher than formerly. It was recommended that these rates be adopted until the city was completely metered, in advance of which it was impossible to state positively the lowest possible rate. During the years from 1901-1905, inclusive, 9,310 meters were placed by the Department at a total cost of $178,746.42, or at an average cost of $19.20 per meter. Of the total number of meters in use 91 per cent. are of the 34 inch size, the balance range up to 6 inches, of which 7 were in use in 1905. The annual costs of repairs and maintenance have naturally varied from year to year, the relative expense increasing as the meters have become older in service. This expense has ranged from 1 per cent. to 6 per cent. of the value of the meters in service, with an average of from 3 to 4 per cent. SUMMARY.-This city, with a population of about 100,000 people, affords an illustration of the effect of practically completely metering all services in three years, and in this respect is unique and instructive. Apparently very careful efforts have been made to detect and stop leakage in the mains and service pipes, and as the greater proportion of all services are metered useless wastage may be assumed to be fairly under control. The results so far are gratifying, although the records do not extend over a sufficiently long term of years to enable forecasts to be made very far into the future. As there are, out of over 10,000, but about 12 consumers using a sufficient quantity of water to entitle them to the manufacturing rate, their combined consumption not aggregating over 4 per cent. of the total supply, it is evident that the per capita consumption should be relatively low 166 Appendix D: Waste Reduction in American Cities. as compared with cities like Milwaukee, where half the total quantity of water supplied to the city is consumed by railways, manufacturing and commercial consumers. Possibly the hope expressed by the Board of Water Commissioners in their annual report for 1901-2, of ultimately reducing the per capita consumption to 50 gallons per diem, may yet be realized. LAWRENCE, MASSACHUSETTS. N 1894, in view of the increasing waste of water, the Water Board appointed a committee to inquire into the advisability of installing meters. This committee reported in October of that year in favor of the system, recommending that meters be placed on all new service pipes supplying dwellings or other buildings where the regular rates exceeded the sum of $5.00 per annum. For the purpose of making the adoption of meters as easy as possible for those who would prefer to use them it was recommended that the city purchase and set the meters, making an annual rental charge of $1.50 per year for one-half-inch meters and in like proportion for meters of larger or smaller sizes, giving the water takers the option of purchasing the meters. The recommendations were not intended to affect the meters already in use, nor water takers paying by fixture rates, unless the Superintendent or agents of the Board found extravagant waste of water, in which case the Superintendent was required to serve a written notice upon the owner or lessee, and if the waste was not checked after a reasonable time he was authorized to attach a meter. The minimum rate was fixed at $7.00 per annum. These recommendations were submitted to the City Council and were immediately adopted and the Water Board authorized to act accordingly. The result was an immediate and marked increase in the use of meters, especially among the class designated as small takers, who found that under the reduced rate they could get their water much cheaper than under the fixture rates. 167 TABLE NO. 32. Average Average Percentage Percentage Population Average Daily Daily of total of Revenue in Families Number of Daily Consumption Number of Number of Consumption Consumption Derived YEAR. and Board- Water- Consumption Per Capita, Services. Meters. Per Water- Passed From inglHouses takers. in Gallons.taker Through Metered Alone. GallonC. Based o n Gallons. Meters. Water Column 2 1877....... 1878....... 1879...... 1880...... 1881....... 1882....... 1883...... 1884....... 1885......., 1886....... 1887...... 1888....... 1889....... 1890....... 1891....... 1892....... 1893...... 1894....... 1895....... 1896....... 1897....... 1898....... 1899....... 1900....... 1902....... 1903....... 1904....... 22,353 26,469 28,391 30,186 31,740 33,304 34,385 34,515 34,718 35,464 36,802 37,436 38,353 39,540 40,852 41,510 43,000 45,216 48,167 50,061 53,613 54,866 57,244 59,534 63,318 65,863 66,603 5,739 6,541 7,001 8,241 8,885 9,043 9,678 10,020 10,233 10,539 10,684 10,848 11,089 11,321 10,244 10,473 10,650 10,804 11,762 12,173 12,685 12,993 13,660 14,339 15,048 15,865 16,050 1,550,000 1,250,290 1,456,029 1,861,363 1,837,553 1,950,658 2,162,919 2,152,675 2,296,265 2,752,665 2,520,115 2,677,259 2,655,000 2,749,000 3,179,150 3,597,420 3,153,000 2,876,543 3,005,624 3,016,928 2,847,869 3,305,000 3,205,000 3,317,000 3,425,425 2,833,683 2,814,933 70 47 50 61 58 59 63 61 66 73 68 71 69 69 78 87 73 63 62 60 53 60 56 55 54 42 42 2,443 2,764.. 3,011 3,224 3,459 3,653 3,896 4,059 4,215 4,356 4,477 4,586 4,713 4,864 4,958 5,128 4,925 5,056 5,216 5,420 5,629 5,777 5,926 6,043 6,375 6,483 6,581 93 138 209 272 332 421 521 609 680 792 918 1,050 1,213 1,402 1,695 2,017 2,231 2,508 2,930 3,339 3,759 4,102 4,386 4,651 5,324 5,603 5,744 270 191 208 226 207 215 223 215 224 261 241 247 239 243 310 343 270 267 255 247 224 254 234 232 227 179 175 No data 14.3 20.7 20.5 22.3 23.4 23.2 26.2 29.0 32.0 34.6 40.0 42.0 37.0 39.0 43.0 42.0 58.0 59.0 No data iss 31. Q 34. 43. 44. 48. 51. a 57.. 59. 60. o 65. 70. 72. 75. 79. 82. 86. 86. 92. Appendix D: Waste Reduction in American Cities. In 1902 the wastage was so apparent that the Water Board early in July ordered the immediate application of meters to all the unmetered services in use. Accordingly about the middle of August meters were set on about 18 business places where the Department had reason to believe water was being wasted in large quantities; the result was the immediate reduction of a considerable atnount of wastage. The policy of placing meters on wasteful consumers has since been continued and the consumption has been reduced about 22 per cent., having been at the rate of 42 gallons per consumer for the years 1903-4 as against 53 and 54 for the years 1900 and 1902. In 1904 about 59 per cent. of the total pumpage was passed through meters and produced 92 per cent. of the total revenue from the sale of water, 87 per cent. of the services being metered. In the reports of the Board for 1902, 1903 and 1904 there is given a distribution of the consumption from which the following table has been prepared. TABLE NO. 33. PER CENT. OF TOTAL PUMPAGE USED FOR DIFFERENT PURPOSES. Not Metered, InMETERED. cluding Some of the Domestic ConYEAR. sumers, Water Total. Used for Public Manufac- Domes- Purposes and turing. tic. Waste. 1902................... 20.3 25.1 54.6 100.0 1903................... 23.0 35.0 42.0 100.0 1904................... 20.0 39.0 41.0 100.0 ~~~..... Assuming that the unmetered domestic consumers, would use the same amount per capita as the metered consumers, if all services were metered, and that 5 gallons per capita per diem would be used for public purposes, an approximate statement of conditions during these three years would be as follows: 169 Waste of Water and Its Reduction. TABLE NO. 34. APPROXIMATE DISTRIBUTION IN LAWRENCE, MASS. Public Uses and a Small Percentage of Domestic Consumption Estimated, in Gallons, Per Capita, Per Day. Percentage Manufac- Domestic Public Not Ac- Not A-g YER. turing Uses. Uses. counted Total. counted Uses. For. For. 1902.......... 11 15 5 23 54 42 1903.......... 10 15 5 12 42 29 1904.......... 8 17 5 12 42 29 SUMMARY.-The use of meters in Lawrence has had a decided and beneficial effect. It will be seen that in 1904 the average daily amount of water supplied to the city was less than in 1895, although the population supplied had practically doubled. It will be noted, however, that during the period that metering was practically voluntary (until 1902) the consumption varied greatly, the wastage not being under complete control. With the placing of a comparatively few meters in 1902, 1903 and 1904 on wasteful consumers selected by inspection, however, a still greater reduction was secured. In explanation of the relatively low rate of consumption per capita, it is to be noted that in Lawrence as in Fall River, a large proportion of the population resides in tenements containing two or more families, as can be observed by the large number of persons per service, about ten, and the comparatively small number of water takers; and under such conditions the number of fixtures from which leakage can take place is much smaller than in cities and towns of a high-class residential character with many detached houses and elaborate plumbing arrangements. Judging by the rate of consumption per capita, and per water taker, the maximum wastage occurred in 1892, during which year at least half the total supply was wasted when judged by the amounts which proved ample in 1904. Going a step further, the analysis of the distribution in 1904 would indicate that there is still about 30 per cent. of the consumption unaccounted for, which proportion may possibly by complete metering and 170 Appendix D: Waste Reduction in American Cities. accounting for all the water used for public purposes be reduced to about 20 per cent., and which would, assuming 5 gallons per capita to be the proper allowance for public uses, correspond with a total consumption for all purposes of about 38 gallons per capita per day. This figure is probably within two or three gallons per dla of the amount to which the consumption of Lawrence can ultimately be reduced so long as present conditions, with respect to manufacturing uses and the character of the city, do not materially change. LO tELL, MASSACHUSETTS. THE rate of consumption of water in Lowell has never been very high as compared with that in many other cities, the highest reached in recent years being 83 gallons per capita )pcr diem in 1900. That there was at least 30 gallons, and probably more, of the 83 wasted, however, has been demonstrated by the records of subsequent years as the result of the policy adopted in 1901 of metering wasteful consumers. Prior to September, 1893, the supply was taken entirely from the Merrimac River, by seepage through a pervious bed of sand and gravel, into a collecting gallery. During the latter part of 1893 an additional supply was put in service, taking water from a gang of driven wells. The ground water supply proving satisfactory, two additional stations were subsequently developed, and by Februaiy, 1896, the city was furnished with ground water only, the Merrimac River supply being discontinued. During the period between September, 1893, and February, 1896, the proportion of river water supplied each year gradually decreased and the proportion of well water increased. Prior to the introduction of this ground water supply Lowell had experienced one epidemic after another of typhoid fever, the death rate from this disease having been extremely high, as can be seen from the following record for the years 1890 to 1896, inclusive: 171. Waste of Water and Its Reduction. TABLE NO. 35. DEATHS FROM TYPHOID FEVER IN LOWELL, MASS., PER 100,000 OF POPULATION. YEAR. 1890. 1891. 1892. 1893. 1894. 1895. 1896. Deaths from typhoid fever......... 158 94 90 66 2 41 40 Percentage of total supply from ground water stations............0 0 0 12 35 67 96 0I It will be seen that the number of deaths each year diminished as the proportion of ground water to total supply increased. During this period the rate of consumption was increasing rapidly, indicating considerable wastage. Of course, the greater the waste the greater the proportion of Merrimac water in the total supply (as the ground supplies were worked to their capacity as rapidly as developed), and hence the greater the danger of spreading typhoid fever. In the annual report of the Water Board for 1895 Mr. R. J. Thomas, the Superintendent, called attention to this possibility, suggesting that the City Councils should adopt legislation of some sort looking to the prevention of useless waste and its resultant evils. Early in 1896 the Water Board determined to place meters on properties where large waste was known to occur, and this resolution, in connection with the public agitation of the meter question, resulted in the applications for over 500 meters during the year. In 1899 half the services in the city were metered, but only about 25 per cent. of the total consumption could be accounted for by the meters. This 25 per cent. of the consumption, however, produced 60 per cent. of the revenue from the sale of water. In the report for 1900 the Board, commenting on the waste of water, expressed their steadfast belief in the metering of every consumer as the only effective means of waste prevention, and determined to make a start in that direction. In the meantime lead poisoning had become more or less prevalent throughout the city, due to the solvent action of the water 172 Appendix D: Waste Reduction in American Cities. from one of the driven well stations, and to reduce its effect as much as possible large quantities of water were wasted from faucets for the purpose of clearing out the poisoned water when drawing water for drinking purposes. After the cause of this trouble was located it was decided to abandon the "Cook" wells and increase the supply correspondingly at other stations. The "Cook" well plant was shut down in May, 1901. During this period the city furnished and set the meters and then invited all those whose consumption, by meter rates, showed less than by faucet rates, to purchase the meters or be placed back on faucet rates. Of the 923 meters set in 1901 there were 377 purchased during the year, while 123 places were put back on faucet rates and 426 remained on meter rates subject to replacement on faucet rates if necessary. The following table shows the population, average daily consumption of water, average daily consumption per capita, number of services in use, number of meters and percentage of services metered. TABLE No. 36. Average Average NUMBER OF PercentPopula- Daily Daily age of YEAR. tion. Consumption Consumption, Meters Gallons. Per Capita, Services Meters to. Gallons. in Use. in Use. Services. 1890 78,000 5,373,536 69 8,732 1,935 22 1895 86,500 6,922,092 80 9,686 3,174 33 1896 88,900 6,933,308 79 9,913 3,759 38 1899 93,000 7,286,205 79 10,362 5,268 50 1900 95,000 7,893,356 83 10,529 5,586 52 1901 97,000 7,059,631 73 10,799 6,609 61 1902 100,000 5,729,325 57 10,984 7,129 65 1903 102,000 5,256,823 51 11,109 7,273 65 1904 104,000 5,485,323 52 11,287 7,513 66 The average number of persons per service has been about 9 during the period covered by these records; which is characteristic of a population, a considerable proportion of which resides in tenements. It will be seen that during 1901 and 1902, when 923 and 520 additional meters were placed in service respectively, these being applied to services of wasteful consumers selected by inspection, such as bar-rooms, tenement buildings, photograph galleries, bot173 Waste of Water and Its Reduction. tling establishments and laundries, the wastage was cut down by about 30 gallons per capita, representing a saving of approximately 36 per cent. of the entire supply. The reports of the Water Board give no detailed information of the relative amounts of water used for public, domestic and trades purposes, and no specific data regarding the cost of metering and meter maintenance. The analysis cannot therefore be carried further than above. SUMMARY.-So long as meters were placed only at the option of the consumers in Lowell, their influence in waste reduction was not felt, even with 52 per cent. of all the services metered; the other 50 per cent., which comprised those heedless or careless consumers who preferred to allow water to waste through leaky fixtures rather than go to the expense of having plumbing put in order, wasted enough, however, to cause the total rate of consumption to increase steadily in spite of the continually decreasing consumption among the economical metered consumers. When the Board of Water Commissioners began to hunt out the worst offenders of this class and put meters on their services, although reaching only 20 per cent. of the number, or 10 per cent. of all the consumers, the effect was felt immediately. It is safe to say, however, that unless this policy is followed out consistently until every consumer is metered, other groups of wasters will probably develop and the original conditions will return. It is only natural that as faucets, ball cocks, etc., become older year by year, they should wear out; and that what are originally small leaks should grow to be larger ones. Unless the plumbing in every city is overhauled occasionally wastage is inevitable and its amount in direct proportion to the carelessness of proprietors in looking after it. This is proven by the experience of every city where control of wastage has been attempted. Ordinarily but a small proportion of the total number of fixtures in use is responsible for the greater proportion of the wastage. When these are found and repaired an immediate benefit results; but such repairs are only effective on the fixtures so repaired, and inevitably others soon get out of order and wastage may be as great or even greater than before. In Lowell the checking of wastage on only 10 or 12 per cent. of the total number of fixtures, selected by inspection, 174 Appendix D: Waste Reduction in American Cities. saved about 35 per cent. of the city's entire supply; whereas prior to the hunting out of these 10 or 12 per cent. over half the total number of services in the entire city were supplied through meters without any apparent effect on the total wastage. MADISON, WISCONSIN.' HE introduction of meters began in Madison in 1887-8. The consumption of water had been increasing at a very rapid rate and threatened soon to reach such proportions that the existing Water Supply Works would be unable to cope with the situation. By 1897 about 90 per cent. of the services were metered and at that time the amount of water supplied to the city per consumer was considerably less than half the amount that had been necessary in 1887-8. At the present time there are only about 100 water takers, out of some 3,500, whose water is not metered, and only about 230, out of the 21,000 inhabitants, who are not supplied with the city water. In 1887 only about half the inhabitants were supplied with city water, and of these only about half of one per cent. were supplied through meters. Thus in the seventh column of the following table it will be seen that although the city is now com-' pletely metered the consumption per inhabitant is much higher than it was in 1887, when there were but few meters; an examination of columns 6 and 8, however, will show the effect of the meters when the consumption per consumer is considered. Column No. 8 was deduced from column No. 6 by assuming an average of 6.06 persons per service, the proportion in 1903-4, and while probably not strictly accurate the ratio between the figures in column 6 and the corresponding figures in column 8 are the same throughout the table, and the comparisons of the yearly reductions in consumption are made more clearly apparent than in the case of the larger figures in column 6. 175 Waste of Water and Its Reduction. TABLE NO. 37. Average >, o 0 0I "O YEAR. Popula- Daily:j 01. W Vo tion. Pumpage, Number INumber i: a P p Gallons. of of c cd ~ * n t < o t V Services. Meters. g n Q*0:g o X. ~.. 003 0+ d m 0SU C 0 1884-5 11,325 546,120 699 3 781 48 129.....5 1885-6 12,063 648,355 876 5 740 54 122....7 1886-7 12,250 715,885 980 5 730 58 121.....6 1887-8 12,500 704,050 1,099 210 640 56 116.. 19.0 1888-9 12,750 535,480 1,229 385 435 42 72...31.0 1889-Oc 13,000 520,030 1,355 441 384 40 63... 32.0 1890-1 13,246 542,162 1,405 498 386 41 64.... 35.0 1891-2 13,900 646,753 1,554 547 414 46 68 41.. 35.0 1892-3 14,600 734,921 1,701 673 432 48 71 48.. 39.0 1893-4 15,300 745,224 1,820 795 409 49 68 49..44.0 1894-5C 15,950 859,467 1,994 1,223 431 54 71 47.. 61.0 1895-6 17,884 891,530 2,186 1,726 408 50 67 44. 79.0 1896-7 18,100 797,186 2,334 2,071 346 44 57 38 14 89.0 1897-8 18,500 750,730 2,473 2,268 303 40 50 36 14 92.0 1898-9 18,800 799,819 2,606 2,410 308 42 51 38 14 92.0 1899-Oc 19,164 837,338 2,758 2,586 303 43 50 34 15 94.0 1900-1 19,500 982,175 2,966 2,807 330 50 55 44 20 95.0 1901-2 20,000 928,246 3,122 2,983 299 46 50 41 21 95.0 1902-3 20,500 1,154,585 3,317 3,176 348 56 57 60 21 96.0 1903-4 21,000 1,501,053 3,462 3,336 433 71 71 52 21 96.0 c Census. Other populations are estimated. ** Based on 6.06 persons per service, the proportion in 1903-4. The figures in this column are not official. They are given to show approximately the ratio of reduction in consumption per consumer from year to year as meterage was extended. The Madison records are of value not only in showing the reduction of wastage on the part of consumers when water was sold by measure, but they show also that where a large proportion of a city's water is used for public and other non-revenue producing purposes the controlling of the wastefulness of consumers alone will not be sufficient to prevent the wasting of a large proportion of the total supply. This is exhibited in the following table: 176 Appendix D: Waste Reduction in American Cities. TABLE NO. 38. PERCENTAGE OF THE TOTAL PUMPAGE TAKEN BY Consumers. Free Water for Schools, Churches, YEAR. Public Buildings, Fountains, Sewer Flushing, Fires, Street Sprinkling, Flushing Mains, Wastage, and Unaccounted for Metered. Not Metered. Water, Not Metered. 1894-5 27.5 58.0 14.4 1895-6 33.0 51.7 15.3 1896-7 30.7 3.9 65.4 1897-8 35.7 2.9 61.4 1898-9 33.2 2.5 64.3 1899-0 36.5 6.1 57.4 1900-1 35.0 1.4 63.6 1901-2 30.4 1.3 68.3 1902-3 27.8 1.0 71.2 1903-4 28.7 1.0 70.3 In the first two years represented in this tabulation the unaccounted for water, including slip of pumps, and loss from leaky service pipes, mains, valves, hydrants, etc., is evidently included in the unmetered water supplied to consumers. In all the data for Madison no corrections have been made for under-registration of meters nor for slippage of the pumps. If it were possible to do this the figures in the second column of the preceding table would be increased and the figures in the last column materially reduced. Of the free water, so called, in Madison the largest amount devoted to any single purpose is for sprinkling macadam streets from May to September, the amount used during the last eight years given in the table being estimated at about 60,000,000 gallons per year. The use of water for public schools, churches and public buildings, including motors therein, has been over half as much as used for street sprinkling. The amount used for extinguishing fires has been about one per cent. of the amount used for street sprinkling. The data regarding these special uses are not given in all the reports, and it is therefore impossible to make up a tabulation showing the percentages used for each purpose each year. It will be seen, however, that since 1897, when about 90 per cent. of the services were metered, to 1904, when 96 per cent. of the services were metered, the consumption of water per metered 177 Waste of Water and Its Reduction. consumer (which means also, practically, per inhabitant) has been only from 14 to 21 gallons per capita per day, while the total per capita consumption has ranged from 44 to 71 gallons per day; the difference, from 30 to 50 gallons per capita per day, representing the water used for public purposes and the unaccounted for water. We have here, therefore, a city in which practically all consumers are purchasing water by meter and in which only about 30 per cent. of the total reported pumpage is accounted for as water used by consumers, a portion of the other 70 per cent. being used for public purposes on a scale unprecedented, and the remainder being unaccounted for. It is apparent that the repeated recommendations of Mr. John B. Keim, Superintendent, that water for these purposes should be metered and the Department given credit therefor, are well founded. In Madison the meters, including the meter boxes, are furnished free to consumers; and in 1888-9, at the introduction of the meter system, the meters were not only furnished, but were also set free of cost to the consumers. Under this system the total amount spent for meters up to September 30th, 1903, was $49,915.57; an average of $15.71 per water taker. The saving in cost of fuel by the reduction of the pumpage per capita is estimated to have repaid the entire cost of the meters and provided a surplus of about $20,000. Water is sold on a sliding scale of charges, the minimum rate being $2.00 for six months' service, for which the consumer is entitled to 1,500 cubic feet of water. The rates for larger quantities are as follows: 1,500- 5,000 cubic feet per 6 months................13 cents per 100 cu. ft. 5,000-15,000 " " " 6 "............... 10 " " 100 " " Over 15,000 " " " 6 "................ 5 " " 100 " " Meters are placed by licensed plumbers at the expense of the consumer. All repairs except when due to frost are made by licensed plumbers engaged by the consumers. Meters broken by frost are repaired for the consumers by the city and the cost collected by the city from the consumer. No rent is charged for the use of the meters except the minimum charge. Meters are read each month, and if the readings 178 Appendix D: Waste Reduction in American Cities. show a larger quantity of water passing through the meters than the normal the owner is notified by the inspector to look out for and repair leaks. MIL WA UKEE, WISCONSIN. N Milwaukee the water consumption per capita had been steadily rising from year to year until in the summer of 1887 it had reached within 5 per cent. of the total pumping capacity of the plant, and within 2 per cent. of the capacity of the intake. The Water Department being without authority and funds to enlarge the capacity of the plant, advantage was taken of a law which had been passed during the previous winter, and meters were placed upon the services of many manufactories and place of business with a view of curtailing wastage and relieving the embarrassment due to short supply and loss of pressure. The result was an almost immediate reduction in the demand on the supply of water, conditions continuing to improve through the Fall and Winter until the diminution had amounted to about 8,000,000 gallons per day, which represented practically 40 per cent. of the average daily consumption for the previous year. It was further observed from the hourly records of the consumption that the greatest proportionate reduction had been in the night consumption, that is, between 11 p. M. and 4 A. M., indicating that the reduction had consisted mainly of the former waste. Previously the night consumption between those hours had been between 70 and 80 gallons per capita, or about 70 per cent. of the average hourly consumption; whereas, after the introduction of about 1,000 meters the night consumption dropped to 50 gallons per capita, or 60 per cent. of the average hourly consumption. The following table gives the population, average daily water consumption, number of services, number of meters in use, average daily consumption per capita, from 1880 to 1904, inclusive: 179 Waste of Water and Its Reduction. TABLE N0. 39. C) 4....a. a,,( *~ * 0. l P. Pp E. - E~ PCT F | 1880 115,587 12,269,000 106 6,888 1,781 26 1881 124,150 13,302,744 107 7,584 1,767 28 1882 132,750 14,690,413 110 8,147 1,803 91 1883 141,350 14,788,701 104 9,133 1,619 221 1884 149,900 14,621,720 97 10,034 1,457 403 1885 158,509 16,062,475 101 10,990 1,461 572 1886 167,700 17,878,436 106 12,212 1,464 871 1887 176,900 20,089,014 113 13,243 1,517 1,728 1888 186,000 19,107,086 102 14,394 1,334 3,523 1889 195,300 19,745,408 101 15,732 1,255 4,749 1890 204,468 22,380,783 109 17,368 1,288 5,876 1891 213,500 23,991,655 112 18,838 1,274 7,526 1892 222,000 22,694,438 102 21,080 1,076 9,008 1893 231,000 24,978,757 108 23,863 1,047 11,347 1894 240,000 25,858,040 108 26,492 976 13,581 1895 249,283 25,291,050 101 29,797 849 15,951 1896 257,000 24,817,029 96 32,118 773 17,930 1897 264,650 23,650,686 89 34,304 689 19,966 1898 272,300 23,188,547 85 35,994 644 22,098 1899 280,000 23,983,022 85 38,191 628 25,639 1900 285,315 23,631,825 83 41,483 569 28,386 1901 300,000 24,069,490 80 43,386 555 31,114 1902 308,000 25,010,800 81 45,480 550 33,815 1903 310,000 26,272,000 85 47,481 553 36,415 1904 315,000 27,940,000 89 49,835 561 39,138 It will be observed that since 1887, when meters were first placed on wasteful consumers by the Department, the daily consumption per service and per capita have been reduced materially and substantially. The consumption per service has been reduced about 67 per cent. and the consumption per capita about 25 per cent. below the figures for 1887. About 78 per cent. of the services are now metered, including practically all the manufactories and a considerable percentage of the domestic consumers. In Milwaukee, as in most cities where they have been introduced, meters were first used by business and manufacturing consumers; and it is noticeable that it was not until ten years after the commencement of the system, by which time a considerable proportion of domestic consumers had been me180 Appendix D: Waste Reduction in American Cities. tered, that substantial benefits in waste reduction had been achieved. The effect of the placing of the first 1,000 meters on wasteful consumers is apparent in the immediate drop from the per capita consumption of 1887, 113 gallons, to 101 gallons the following year. These meters were too few, however, to accomplish lasting results, waste having developed in other sections of the city where no meters had been placed. With the extension of the meter system the number of consumers who could waste with impunity was gradually reduced until their number was too small to seriously affect the average consumption of the entire city. It will be seen that during the ten years from 1891 to 1901 there was practically no increase in the total average amount of water used daily in Milwaukee, although the population of the city had increased from 213,500 to 300,000 and the number of water takers had increased 130 per cent. in the same time. In 1898 about 70 per cent. of all the services were metered. This included water supplied to all railroads, manufactories and business houses in the city, as well as that furnished for power purposes, yet the amount of water metered was not over 50 per cent. of the entire supply, the other 50 per cent. being taken mainly by unmetered domestic consumers, among whom there were still many who were extravagantly wasteful. During February, 1899, the meter rates and flat rates were readjusted so that all consumers, domestic as well as manufacturing, should pay a uniform price for the water they consumed. No minimum charge was set, but the price was carefully adjusted so that sufficient revenue would be produced to maintain the plant in successful operation. This rate, for all metered consumers, was fixed at the uniform price of 4 1-2 cents per 100 cubic feet, or 6 cents per 1,000 gallons. In addition to this there was a fixed charge of $1.00 for reading each meter, which in effect constitutes the minimum rate for metered water. Based upon the metered consumption of 1898 there were 12,000 metered consumers whose annual tax for water, including the charge for reading the meter, was less than $3.00 and 5,900 such consumers whose annual rate was less than $2.00 per year, a rate unparalleled by any city in the country. 181 Waste of Water and Its Reduction. The action establishing this low rate has met with popular endorsement, the average number of meters voluntarily installed yearly since 1898 having been about 2,700 as against 2,300 for the previous six years and 1,500 for the six years before that; 3,541 were installed in 1899, which was the largest number placed in any year since the establishment of the Water Works. A great many domestic consumers availed themselves of this rate. The meters are read by a uniformed force of meter readers, the uniform being a protection to the consumer and to the meter readers as well. The city does not, excepting in a very few cases, furnish nor maintain the meters, but since 1899 has repaired all meters upon request, charging only the actual cost of the work. No specific data therefore are available as to the cost of metering the city, or as to the cost of maintaining the system. SUMMARY.-The general introduction of meters in Milwaukee has been of undoubted and great benefit in curtailing waste. It is to be noted, however, that 70 per cent. of the consumers, using about 50 per cent. of the entire supply, are railways, manufacturing and business concerns. This fact may influence to a considerable extent the total wastage and tend to make the percentage of waste to total supply smaller than in cities where the manufacturing uses are less extensive, since metered manufacturing consumers do not waste water. In this case, therefore, the wastage is limited to the 50 per cent. of the supply taken through the 28 per cent. of unmetered domestic water services. Since the reduction of the meter rates to a uniform basis there has been a voluntary extension of the meter system to all classes of consumers, and it seems likely that this attitude towards meters will be maintained. The reduction of waste has not yet reached the attainable limit, although with the waste entirely stopped on half the supply, the waste on the other half, representing about 28 per cent. of the services, does not seem to have increased proportionately for several years, notwithstanding the rapid growth of the city. The introduction of meters has brought about a permanent and lasting improvement in suppressing waste. Steadily, with the ex182 Appendix D: Waste Reduction in American Cities. tension of the meter system, the wastage has been curtailed each year in sufficient quantity to provide for the increase in population for eleven years; and since 1901 the increase in the quantity of water required has been about in the same proportion as the increase in the population. Since 50 per cent. of the supply in Milwaukee is used for manufacturing and business purposes, approximately 45 gallons per capita per day represents the domestic and public uses, the wastage, the leakage from the street mains and the unaccounted for water. It would not seem, therefore, that a much greater reduction in wastage is to be looked for. These records show clearly the folly of assuming that since certain foreign cities get along on 15 gallons per capita per day, and certain American cities on from 25 to 60 gallons per day per capita, all cities should be able to do the same. The total amount of water used daily for manufacturing purposes alone in Milwaukee, a city of 320,000 persons, is half the total consumption for all purposes, domestic, public and manufacturing, of Berlin, a city of 2,000,000 inhabitants; and the manufacturing consumption per capita in Milwaukee three times the per capita consumption for all purposes in Berlin. NEWARK, N. J. INCE 1892 Newark has been supplied by gravity with water from the conduits of the East Jersey Water Co., the water being delivered into the city's reservoirs through Venturi meters. Under the original contract with the East Jersey Water Company the city was entitled to 27,500,000 gallons daily, which was thought to be sufficient to provide for all legitimate wants for several years. At that time about 1,000 meters, owned and set by consumers, were in use on the service pipes of manufacturing and business establishments. As early as 1895 it was found that during cold weather, when consumers had allowed the water to run from faucets all night to prevent the freezing of the service pipes, the rate of consumption had on certain days exceeded the allowance under the contract. 183 Waste of Water and Its Reduction. The Engineer of the Department, Mr. M. R. Sherrerd, mentioned this in his annual report for 1895 and suggested the general introduction of meters on all service pipes, except those for domestic consumers, as being likely to hold the daily average below the maximum allowance for a few years, possibly until 1900. He also suggested the enforcement of a penalty for waste, or a provision for placing meters on services where waste to prevent freezing was known to take place, and further recommended more stringent regulations regarding plumbing, both as to its character and as to the proper locations of pipes to avoid the risk of their freezing. In 1896 an auxiliary supply of ground water was drawn upon, and on one occasion it was necessary to have recourse to pumping from the Passaic River in order to keep up fire pressure. In the Winter of 1897 ground water was again pumped during parts of December, January and February, and the average lafly consumption increased about 2,000,000 gallons over that of the previous year. During 1898 the consumption had increased so greatly that the pumping plant was kept under steam practically the entire year and a total of about 191 millions of gallons of water was drawn upon from the ground water plant to keep up a sufficient supply for the city's needs. During the Winter of 1898-99, which was unusually severe, 146,600,000 gallons were pumped from the ground water station, 63,000,000 gallons of which was a mixture of ground water and Passaic River water, as the water from the East Jersey Water Company's conduits and from the ground water station could not together supply enough for the demands of the city. As a result of the shortage of water during the blizzard of February, 1899, the Board of Street and Water Commissioners decided to undertake at its own expense the general introduction of water meters, to curtail the waste. Accordingly, during that year inspections were made to locate wasteful consumers and a total of 3,749 meters were set by the city, 1,334 additional meters having been bought and set by different consumers. The result of this action, coupled with the relatively short durations of cold spells during the Winter, had the effect of materially reducing the daily consumption. During the succeeding year the Department placed 1,352 meters and the consumers 1,360, a total of 2,712, and 184 Appendix D: Waste Reduction in American Cities. the consumption of water was still further reduced, as shown in the following table: TABLE NO. 40. Numbr of M s. Average Daily Consumption Number of Meters Set. in Million Gallons. MONTH. ________ 1898 1899 1900 1898 1899 1900 January...... 21 40 524 28.2 29.9 25.7 February..... 17 18 338 28.0 32.7 25.5 March......... 30 47 294 25.1 27.7 24.4 April......... 20 117 351 25.3 26.2 22.7 May.......... 133 714 496 25.4 27.1 22.8 June........ 62 627 176 27.5 27.1 23.7 July.......... 54 259 82 28.4 26.3 24.1 August....... 49 363 86 29.4 25.5 24.4 September.... 34 341 77 28.5 25.3 24.7 October......... 80 439 97 27.6 25.8 24.1 November..... 124 545 135 27.1 24.8 23.2 December...... 58 673 56 28.0 24.2 23.5 Total.. 682 4,083 2,712 27.2 26.8 24.1 Up to 1898 the normal annual increase in the daily average consumption had been about 2,000,000 gallons, but from 1898 to 1900 there was a decrease of 3,000,000 gallons per day instead of an increase of 4,000,000 gallons, a credit of 7 million gallons per day reduction from what might have been expected under normal conditions. The following table exhibits the daily water consumption, number of services, number of meters in service and daily average consumption per capita in Newark from 1870 to 1904, inclusive: 185 Waste of Water and Its Reduction. TABLE NO. 41. AVERAGE DAILY CONSUMPTION, Average Daily Nmber Number IN GALLONS. YEAR. Population. Consumption, of of Gallons. Services. Meters. Per Per Capita. Service. _Number_ 1870........ 105,542 1,440,000 3,383...... 14 445 1871........ 109,500 2,479,000 4,480...... 23 550 1872........ 114,000 3,571,000 5,616...... 31 640 1873......... 117,000 4,473,000 6,598...... 38 680 1874......... 120,000 4,732,000 7,389...... 39 640 1875........ 123,310 5,430,000 8,195...... 43 650 1876........ 125,250 6,206,000 8,885...... 49 700 1877......... 127,400 6,714,000 9,485...... 52 708 1878........ 130,800 7,280,000 10,048...... 55 726 1879........ 133,600 8,375,000 10,557...... 62 794 1880....... 136,508 9,200,000 11,093...... 67 828 1881......... 139,000 9,888,000 11,902...... 71 830 1882........ 142,000 9,680,000 12,676...... 68 754 1883........ 145,000 9,672,000 13,592 187 66 712 1884........ 148,500 10,504,000 14,649 267 71 718 1885......... 152,988 11,400,000 15,908 262 74 716 1886........ 158,000 11,784,000 17,111 292 74 688 1887......... 163,000 12,642,000 18,410 353 77 686 1888......... 168,500 13,531,000 19,854 401 80 682:1889........ 174,500 14,079,000 21,532 456 80 654 1890........ 181,518 15,260,000 23,265 526 84 657 1891......... 187,000 16,630,000 25,007 615 88 665 1892........ 193,000 17,732,000 26,455 1,075 91 6'71 '1893......... 200,000 20,000,000 27,565 1,257 100 726 1894......... 207,500 20,040,000 28,537 1,397 96 702 1895......... 215,000 22,100,000 29,646 1,660 102 746 1896.. 225,000 22,779,000 30,792 2,073 101 740 1897......... 234,000 24,969,000 31,686 2,577 107 788 1898........ 242,000 27,400,000 32,537 3,259 113 843 1899......... 250,000 26,800,000 33,494 7,342 107 800 1900......... 258,000 24,066,000 34,353 10,054 96 702 1901....... 265,000 24,900,000 35,284 11,502 94 706 1902........ 273,000 26,100,000 36,278 13,274 95 720 1903......... 282,000 26,610,000 37,180 15,019 94 718 1904......... 290,000 30,553,000 38,185 16,026 105 800 A study of these figures shows a steady and rather uniform increase in population from year to year. In the 25 years from 1870 to 1895 the population doubled, but the quantity of water used increased 16-fold, and the number of services nearly 9-fold. In earlier years the proportion of water takers to population was as 1 to 30, but this ratio has been steadily decreasing until in 1904 the ratio was one service to every 7.5 persons of the resident population, indicating that at the present time probably about 80 to 85 186 RESULT OF THE COMPULSORY INTRODUCTION OF METERS IN NEWARK, N.J., DURING 1899 1900 DIAGRAM FROM ANNUAL REPORT FOR 1-903 OF STREET AND WATER COMMISSIONERS, NEWARK, N.J. )3 1894 1895 1896 1897 1898 1899 1900 1901 1902 1 aI w I I >r: W tL _J (n z -I _J L5 I I I u(/ 0 -UO 0 (5 FZ W U 0 -i I I < --- —— ALL METERS SET BY CONSUMERS ----------— 4116 METERS SET BY-> —ALL METERS SET BY CONSUMERS —* CITY ON WASTEFUL CONSUMERS SELECTED BY INSPECTION r '1 r I Appendix D: Waste Reduction in American Cities. per cent. of the reported population is supplied with the city water as against about 14 per cent. in 1870. The low per capita consumption in the early years and the slow but regular increase from year to year indicates, therefore, merely that there were relatively more people supplied with water each year, not that less water was consumed per user. This is shown by the average annual consumption per tap, which since 1872 has not, with the exception of eight years in the 34, varied more than 120 gallons, or about 14 per cent., from highest to lowest. The period which is of greatest interest during the 35 years covered by these statistics is that from 1894 to 1904. From 1894-98 the average daily consumption increased about 37 per cent. and the consumption per service about 20 per cent., while the population increased but 16 per cent. Two hard winters contributed to this effect, owing to the large amount of water wasted to prevent the freezing of pipes in the houses, but still the increase was abnormal and alarming. The placing of the meters by the Department had a decidedly beneficial effect. They were placed on the services of consumers who were found to be wasting water needlessly and carelessly. Only 4,116 meters were set by the Department during the two years. By an inspection of the accompanying diagram the effect of the introduction of these meters can be readily seen. No additional meters have been set by the Department since September, 1900, yet those placed have satisfactorily held the rate of consumption down approximately to the point reached at the time the Department ceased placing them. A decided difference is to be noted between the effect of the meters placed by the Department and those placed by the consumers, the latter having comparatively little effect in reducing the rate of consumption. The explanation of this difference is that the meters placed by the Department were purposely set on the services of consumers who were notorious wasters, while consumers who set their own meters cannot be supposed to have done so to punish themselves for wasting water, but rather to avail themselves of the privilege of paying for what they actually used and thus escaping the burden of helping to pay the cost of the extra water wasted by the careless consumers. 187 Waste of Water and Its Reduction. At the present time about 42 per cent. of the services are metered and the normal per capita consumption is about 95 gallons per day. Undoubtedly among the 48 per cent. of unmetered consumers, mostly domestic, there are still a great many careless wasters, and as their number becomes smaller the difficulties of maintaining an adequate pressure and supply will become less. The increase in consumption per tap to be noted in 1904 was caused by the very cold weather in January, February, March and December, the total wastage to prevent the freezing of service pipes amounting to approximately 660,000,000 gallons. SUMMARY.-The compulsory placing of meters on 4,116, or 10 per cent. of the total number of consumers, in 1899 and 1900, these 4,116 being selected by inspection, was followed by a reduction of wastage to the extent 'of three million gallons per day, or 11 per cent. of the supply. During 1901, 1902 and 1903 this advantage was held, the consumption increasing only in proportion to the population. In 1904 the consumption again increased abnormally, due in part to wastage during the severe winter weather. The increase (luring the Summer of 1904 was largely chargeable to the lack of rain, the precipitation from June to October, 1904, being only 24.2 inches as compared with 48.1 inches in 1903, this circumstance giving rise to a much larger use of water than usual for washing pavements and sidewalks and for sprinkling streets and lawns and for gardens. On the whole, however, the consumption was too high to be satisfactorily accounted for in this way; and it is probable if not certain that another class of wasters has sprung up among the 48 per cent. of domestic consumers not metered. PROVIDENCE, RHODE ISLAND. SINCE 1872 meters have been in use in Providence. By 1877 42 per cent. of the services were metered and from that date the number of meters set has increased more rapidly than the number of services added, until in 1904 86 per cent. of all services were delivering water through meters. During this period, from 1877. to 1904, the population of the city has increased 102 per cent., while the average per capita consumption has increased 178 per cent. or from 24 to 67 gallons 188 Appendix D: Waste Reduction in American Cities. per day. This increase, it will be noted, has taken place in spite of the extensive use of meters. About 1890 the consumption began to increase, and in 1892 the Department started an investigation to find out the reasons therefor. In a paper by Mr. J. H. Shedd, M. Am. Soc. C. E., read before the New England Waterworks Association in March, 1904, the results of this investigation were described very clearly. The following table, abridged somewhat from the one given by Mr. Shedd, and extended to include the records of two more years, shows the population supplied with water, the number of miles of street mains per 1,000 of population, the per capita supply of water and percentage of services metered from 187T to 1904, inclusive: TABLE NO. 42. DISTRIBUTION, POPULATION, AND PER CAPITA SUPPLY OF WATER IN PROVIDENCE, RHODE ISLAND. Miles of Per Capita YEAR. Miles of Population Pipe Per Supply of Percentage of Street Mains. Supplied. M. of Water, Services Metered. Population. Gallons. 1877..... 143.921 102,900 1.398 24.21 42.00 1878..... 149.066 104,800 1.422 26.00 43.00 1879..... 151.389 106,875 1.418 29.00 45.00 1880..... 154.566 109,400 1.413 32.00 47.00 Mch., '80 1881..... 159.404 112,200 1.421 33.00 48.00 Nov., '80 1882..... 166.720 115,100 1.446 32.00 No record 1883..... 176.676 118,000 1.497 35.00 52.00 1884..... 183.839 120,900 1.521 34.00 54.00 1885..... 189.304 123,800 1.529 38.00 55.00 1886..... 194.230 126,700 1.533 38.00 57.00 1887...... 203.389 129,700 1.568 38.00 58.00 1888..... 210.367 132,600 1.586 42.00 59.00 1889..... 218.010 135,700 1.607 43.00 61.00 1890..... 226.479 138,700 1.633 49.00 62.00 1891..... 243.532 141,800 1.717 51.00 63.00 1892..... 257.715 144,900 1.779 56.00 65.00 1893. 273.265 148,000 1.846 63.00 69.00 1894..... 284.760 151,800 1.876 65.00 72.00 1895..... 296.275 156,400 1.894 57.00 75.00 1896..... 303.192 162,200 1.869 56.00 76.00 1897..... 310.346 168,200 1.845 51.00 78.00 1898..... 314.853 174,200 1.808 52.00 80.00 1899..... 318.428 180,600 1.763 53.00 81.00 1900..... 324.555 186,800 1.737 54.00 83.00 1901..... 331.035 193,000 1.715 56.00 84.00 1902..... 336.156 199,400 1.686 58.00 84.00 1903..... 341.400 202,800 1.684 65.00 85.00 1904..... 345.710 207,900 1.662 67.00 86.00 189 Waste of Water and Its Reduction. The following is quoted from Mr. Shedd's paper, referred to above: "In the year 1892 the Department noticed what was considered to be an unnecessary per capita use of water, and special inspectors were engaged to examine the supply through services, and the returns of the ordinary inspection were reviewed in the office to discover a reason for such use. About two years were occupied in the examination and in applying remedies. It was found that upon many of the unmetered services there was large waste, and the takers on such services were notified that under the provision in the published rates allowing special assessments to be made for peculiar circumstances they would be subjected to such special assessment, unless they chose to put on meters. A considerable number of such takers had meters set and the rate of increase in the number of meters showed the influence of this move. The examination and the effort to check waste reached its culmination in 1894. In that year the average daily consumption of water was 9,904,434 gallons and the per capita supply was 65.24, which was larger than it had been before or has been since, and undoubtedly more than was reasonable. The effect of the special effort to check waste was shown in the following year in two ways that were striking when observed upon a profile illustrating the conditions obtaining upon the work. First.-The total average daily consumption of water dropped to 8,905,085 gallons and the per capita supply to 56.93, which was still fully up to the requirements for use. Second.-The rate of increase in the revenue shot up in a marked way. The full effect of the effort to check waste was not, however, shown until 1897, when the total average daily consumption of water had dropped to 8,635,067 gallons and the per capita supply to 51.34. "From that time to within about two years the increase in the total supply has been reasonably consistent with the increase of the population supplied. "Lately a new tendency is shown to dispose of a greater per capita supply of water. This seems to be accounted for by the effect of electrolysis which has a destructive action in several places in the City. On one street several street mains have been taken out in two successive years, being badly decomposed and one of which had been completely eaten through. The iron bolts of meters have been eaten off and replaced by composition bolts. Lead service pipes have also been badly eaten. The conditions favoring the destructive action of electrolysis are being changed steadily for the better, and it is hoped that the waste of water from this cause may be kept within reasonable limits. "It is believed there is a considerable waste in connection with the.direct service to elevators from the pressure in the mains. One hundred 190 Appendix D: Waste Reduction in American Cities. and sixty-nine elevators are supplied in this way and the condition of their supply is now being examined. "The working force of the Meter Department in Providence consists of three men who do all the setting and repairing of the meters on the system. The total expense of the Department is more than returned to the City by the charges against the takers, they being at a fixed rate per meter for the usual operations. The average anuual cost per meter for repairs is about eleven and one-half cents, the total cost for the last full year reported being $2,278.89. "Eternal vigilance is as necessary in a Water Department as in other matters, to secure a perfect result. It is found that in Providence there are men so lost to the dictates of a good conscience that they will tap in a supply back of the meter so as to draw water without having it measured. "Water is supplied to manufacturing establishments for fire purposes without cost when contained in pipes laid at the expense of the establishment and not connected with any other means of supply. It is a condition of such supply that no water whatever shall be drawn from these pipes except to extinguish fires. It was found at one time that about a million gallons per day extra was being drawn in a manufacturing section of the City and an Inspector was sent out to locate the waste. It was found that two establishments near each other had opened their fire service pipes and were using the water for washing and other purposes. The fact that the combined use of the two was so great facilitated the discovery. "Other surreptitious use of water has been discovered from time to time, like its use for broad irrigation, etc. "Nearly all the evils of waste can be cured by placing a meter on every service. It would hardly be expected in any community that illuminating gas would be furnished to takers except through a meter, notwithstanding the fact that gas is cheaper than water. In Providence gas is sold at $1.10 per thousand cubic feet, while for water $1.50 per thousand cubic feet is paid." The average cost per year of maintaining the meters in use for 1878 to 1886 inclusive, for each meter in use, is given by Mr. E. B. Weston as $0.44 (the condemned meters not being included) and $0.78 per meter per year, including the cost of the condemned meters. From the quarterly reports of the Board of Public Works it would appear that the average cost of the meters set between 1899 and 1905 was $16.50 per meter (all sizes included), and the cost of maintenance and repairs about 2 per cent. of the value of the meters in service. 191 Waste of Water and Its Reduction. SUMMARY.-These statistics of a city which is very largely metered, and which has been held up for many years as an example of a city in which the consumption has been kept down to about 50 gallons per day, show in recent years an increase which meters have not been able to control. This increase has been attributed to the destruction of service pipes by electrolysis. The only likely explanations are either, as suggested, that the extensive pitting of the pipes is causing the leakage or that the uses for manufacturing, public and unclassified purposes are increasing more rapidly than the general demand for domestic consumption. The data at hand are not sufficient to demonstrate the extent to which the demands for these purposes are influencing the total supply. Until within late years the rate of consumption has kept pace in about even proportions with the increase of population, although undoubtedly some influence other than meters has, particularly in the earlier years, contributed to this result. It is not conceivable that the checking of waste on 50 per cent. of the consumers could have kept the consumption so constant and at so low a figure. This Water Department has always been very carefully managed and wastage has been kept down by a published and well understood warning that those who waste water would be subject to a special assessment. RICHMOND, VIRGINIA. TIE introduction of meters at the expense of the city for the purpose of preventing wastage was begun in 1897 and has been continued with more or less regularity since that time. During the cold weather of 1893 so much water was wasted to prevent the bursting of pipes that the city reservoirs were practically emptied and there was little or no pressure in the mains in the principal streets of the city. Repetitions of this condition finally led to the use of meters to control this excessive wastage, and since metering began no such troubles have been experienced. The effect of metering, however, has not been sufficiently marked up to the present time to indicate any material reduction in the amount supplied per capita, which is high for a city of the size and character of Richmond. 192 Appendix D: Waste Reduction in American Cities. TABLE NO. 43. AVERAGE DAILY CON- Per SUMPTION, IN GALS. Total cent. of No. of Popula- No. of Ser- Miles of Meters YEAR. tion. Meters. vices Mains. Owned Per ters. Me- by City. Total. Capita.* tered. 1899..... 83,000 10,048,549 126 4,296 30 92.72 2,443 1900..... 85,000 10,950,000 129 4,621 32 94.07 2,716 1901..... 88,000 10,999,893 125 4,978 34 97.29 3,045 1902..... 92,000 11,246,899 122 5,609 37 103.30 3,631 1903..... 96,000 11,841,435 123 6,201 40 105.89 4,291 1904.... 100,000 12,905,342 129 6,653 41 108.02 4,875 *Based on estimated population stated in second column. It is estimated that at the present time, March, 1906, about 75 per cent. of the domestic consumption is metered and about 45 per cent. of the total consumption. The cost of furnishing and setting the meters owned by the city is about $12.00 per meter; the cost of setting meters for the consumers averages about $2.62 per meter. The total amount spent for meters from 1897 to 1904, inclusive, was $50,000, which for the 4,875 meters in service represents an average cost of $8.37 per meter; this, however, makes no allowance for the meters in stock, adding which would reduce the average cost somewhat. Figures on the cost of maintenance and repairs are not available. Regarding the effect of the introduction of meters Mr. Charles E. Bolling says: "In 1897 nearly 500 houses had no water in the second stories during the day. There are now no complaints from these residents. The pressure at the high points is three times greater than formerly. "The pressure is still improving, the revenue increasing and the popularity of the method of paying by measure shown by the number of applications for meters. I regard it as most important that all city buildings should have meters. The new city jail consumes 2,000,000 gallons of water per month, which if paid for at 5 cents per 1,000 gallons would amount to $1,190 per annum, whereas the amount actually received is but $54 per annum. The average consumption per capita at the jail is 500 gallons per day. The pressure on the mains has been maintained through periods when it was 193 Waste of Water and Its Reduction. very cold, and to a considerable extent has controlled the waste. Complaints of high bills, when investigated, invariably disclose leaky pipes or fixtures or careless waste." From the data at hand it would appear that the chief benefit from the introduction of meters has been the prevention of some of the waste during cold and during hot weather, with consequent increase of pressure in the high districts. Apparently a large amount of wastage continues from the 55 per cent. of consumption not metered, which includes 25 per cent. of the domestic consumption, all public consumption and some of the manufacturing and business consumption. So far as the reduction of the per capita consumption is concerned apparently no benefit has been derived. Possibly the meters have not yet reached the worst offenders. Yet it is worthy of note that in this case the metering of 75 per cent. of the domestic consumers has not materially affected the per capita consumption of the city. ST. LOUIS, MISSOURI. T. LOUIS began to combat wastage by house-to-house inspections in 1880, the agitation having been started in 1878-9 and 1880 by recommendations of the Board of Public Improvements looking to universal metering. This recommendation, however, was never adopted, and to this day meters have only been very sparingly introduced. Such efforts as have been made at waste reduction in St. Louis have therefore been confined entirely to house-to-house inspections conducted with vigor at times and neglected at other times. The net result has been a steady tendency toward an increase in the rate of consumption, with an occasional reduction for a year or two at a time, as indicated in Table 44 on opposite page. The result of this method of checking waste in St. Louis is summed up by Mr. M. L. Holman, in a paper read before the Engineers' Club of that city, and published in the Journal of Associated Engineering Societies, August, 1885. Of about 100 house inspections made by Mr. Holman on the night of July 31, 1884, 66 per cent. were found to be all right and 194 Appendix D: Waste Reduction in American Cities. TABLE NO. 44.; YEAR. Population. 1881...... 1882...... 1883...... 1884...... 1885...... 1886...... 1887...... 1888...... 1889...... 1890...... 1891...... 1892...... 1893...... 1894...... 1895...... 1896...... 1897...... 1898.... 1899...... 1900..... 1901...... 1902...... 360,600 370,800 380,900 391,000 401,100 411,270 421,400 431,500 441,600 451,770 464,100 476,500 488,800 501,200 513,500 525,900 538,200 550,500 562,900 575,238 588,000 601,000 Average Daily Consumption, Gallons. 27,500,000 28,000,000 27,400,000 24,600,000 26,800,000 30,000,000 30,500,000 32,000,000 32,300,000 35,200,000 38,600,000 42,100,000 47,500,000 52,900,000 55,200,000 51,100,000 52,600,000 54,400,000 60,800,000 63,500,000 65,800,000 67,000,000 AverDaily Consumption per Capita. 75 74 72 63 67 73 72 74 73 78 83 88 97 105 107 97 98 99 97 111 112 112 20,204 21,745 23,648 25,321 27,457 29,884 31,794 34,022 36,082 38,183 41,331 44,382 47,445 50,540 53,354 56,865 59,423 61,839 63,851 65,688 67,243 69,483 Number No. of of MeServices. ters. 573 905 1,228 1,522 1,811 2,143 2,376 2,662 2,888 3,115 3,399 3,601 3,750 3,850 3,979 4,092 4,198 4,151 4,161 4,133 4,331 4,525 Percentage of Services Metered. 2.8 4.2 5.2 6.0 6.6 7.2 7.5 7.8 8.0 8.1 8.2 8.1 7.9 7.4 7.5 7.2 7.0 6.7 6.5 6.3 6.5 6.5 Average Daily Consumption per Service. 1,360 1,290 1,160 970 975 1,005 960 940 894 920 935 950 1,000 1,050 1,035 898 885 880 952 965 978 965 I 34 per cent. were found to be wasting water; the day gang, which followed up the night inspections, found that of the 34 per cent. of the above houses found to be wasting water on the previous night 62 per cent. had defective plumbing. The following quotation is from the paper by Mr. Holman, referred to above: "There can be no doubt but that by the employment of these inspectors a large waste of water has been stopped. I doubt, however, if as good results can be expected to follow a continuous practice of this system. Indeed, I am more than ever convinced from my observations of the work done by this force, that the only permanent and economical method of controlling the delivery of water to consumers for other than business purposes is in operating under some law similar to the proposed ordinance sent to the Municipal Assembly by the Board of Public Improvements. "Although the saving of water was considerable by this system of inspection, yet during the two months that I had charge it became 195 Waste of Water and Its Reduction. very onerous to the water-takers. I was bothered by a great many groundless reports against my inspectors, representing them guilty of gross misconduct, etc., etc. Mr. S. F. Burnet and myself did nearly all the key work, and I generally found several reports on my desk during the day, giving us a good overhauling. Some persons thought to defeat the inspection by filling the stop-boxes full of sand; others tried old shoes, hard coal well rammed down, etc. "I regard the night inspection system as a very good method of house-to-house inspection, as it puts only those who are found running water to waste at night to the inconvenience of a day inspection. The great drawback is that a waste of water cannot be proven unless found by a day gang of inspectors employed to follow up the night men, all sorts of excuses being offered to account for the flow of water through the stop-cock during the night hours. "The vital point of any method of house-to-house inspection is that the inspection can only be made efficient by keeping a very expensive force of inspectors, and they must be backed up by complete legislation. That the benefits of the inspection are only temporary is, I think, fully shown by the detail reports given." SPRINGFIELD, MASSACHUSETTS. I N Springfield meters have been in service to a limited extent for a good many years, but as their use is not compulsory, those now in place have been set at the request of economical consumers for the purpose of reducing their water bills and not for the purpose of reducing wastage. The statistics of water consumption in the town show this condition very distinctly, as may be seen from an inspection of the following, Tables 45 and 46 (page 197). There is so much uncertainty regarding the total quantity of water used at Springfield, the quantities given in the table being estimated by the Water Commissioners (not exact measurements), and so little detailed information is to be had regarding the amounts used for different purposes that only very general deductions are possible. However, as a considerable proportion of the consumption is metered, it is possible to show in a measure proportional to the actual conditions that the increase of consumption and the irregular fluctuations thereof are probably confined almost entirely to the unmetered portion of the water. From the first table it will be seen that the consumption per inhabitant as well 196 Appendix D: Waste Reduction in American Cities. TABLE NO. 45. AVERAGE DAILY CON- NUMBER OP SUMPTION. Per - Per- Average No. of Average centage centage No. of Con- Daily Con- of Me- of total ConYEAR. sum- sumption, Per Per ters to Con- sumers ers. Gallons. Con- In- Servi- Me- Servi- sump- per sum- habi- ces. ters. ces. tion Service. er. tant. Metered. 1893 37,000 4,320,000 117 86 7,000 1,682 23.9 17.2 5.30 1894 38,500 4,524,510 117 87 7,482 1,947 26.0 17.0 5.12 1895 41,000 4,638,060 113 90 7,900 2,207 27.9 21.0 5.19 1896 42,500 4,819,828 113 88 8,298 2,299 27.6 21.2 5.12 1897 44,000 5,000,000 113 88 8,721 2,496 28.6 19.7 5.04 1898 45,500 5,142,500 113 89 9,070 2,727 30.0 21.6 5.02 1899 47,000 5,307,000 113 87 9,456 2,898 30.6 20.0 4.97 1900 48,200 5,471,500 114 88 9,764 3,122 32.0 21.2 4.94 1901 50,000 8,000,000 160 129 10,034 3,337 32.3 14.9 4.98 1902 51,000 8,300,000 162 128 10,221 3,537 34.6 15.4 5.00 1903 54,000 11,700,000 216 176 10,471 3,798 36.2 12.8 5.16 1904 74,916 9,720,000 130 128 10,641 3,987 37.5 17.9 7.04 TABLE NO. 46. AVERAGE DAILY CONSUMPTION, AVERAGE DAILY CONSUMPTION GALLONS. PER CONSUMER.t Average Daily ConYEAR. Stion per Per Unmetered YEAR. Service for Per Metered Ser- Service, Including all Services, Per Per Un- vice, Including Part of Domestic Gallons. Meter. metered Manufacturing and and All Public Service. Part of Domestic Uses, and Unac* Consumption. counted for Water. 1893........ 617 441 813 83 153 1894......... 604 394 817 77 160 1895........ 587 442 816 85 157 1896........ 581 444 802 87 157 1897........ 573 394 803 78 159 1898........ 567 407 811 81 161 1899........ 561 366 810 74 163 1900........ 560 373 823 76 167 1901........ 797 355 1,196 71 240 1902........ 812 360 1,243 72 248 1903........ 1,117 397 1,754 77 340 1904........ 913 438 1,462 62 208 *Number of unmetered services assumed to be total number of services less the number of meters in use. tNumber of persons per unmetered service is assumed to be the same as per metered service, and in both cases was obtained by dividing total number of consumers by total number of services in use for corresponding year. 197 Waste of Water and Its Reduction. as per consumer is high, in spite of the metering of about 37.5 per cent. of the services, which percentage includes a portion of the domestic consumers, possibly about one-third the total number. There has been maintained for many years a bureau for the inspection of the fixtures in houses not metered, and considering the limited number of inspections a large number of leaks have been discovered and stopped each year. No very definite conclusions can be drawn from these inspections; they are made when the inspectors can take the time for them from other duties, and hence do not as a rule cover the entire city each year. In general terms it is safe to say that leaks are now more frequent and in greater proportion to the number of fixtures in use than in former years. In 1893 the general average of leaks found and stopped was about 4 to a day's work of an inspector, while in 1904 the general average had increased to about 6 leaks per day's work. During this same period the number of water closets had increased from 9,768 to 20,171 and the number of bathtubs from 4,270 to 10,823. The ratio of water closets to population has remained fairly uniform at about 3.7 persons per closet, while the ratio of persons to bathtubs has decreased from 8.6 persons per tub in 1893 to about 5 or 6 persons in 1904, thus indicating a general increased use of more sanitary conveniences with the increasing prosperity of the city during this period. In spite of these inspections, which are more systematically carried on than in many cities and which for several years have resulted in the stopping of nearly 1,000 leaks a year, their effectiveness when measured by permanent results, or even by the reduction in general wastage, is inconsiderable. Using the recorded amounts of water passed through meters, with the population and number of services for each year, as a basis for estimating the average number of persons per service, the relative amounts of water used per day, through the metered and unmetered services, and the amounts per capita, deduced as above indicated, would be approximately as shown by Table 46. 198 Appendix D: Waste Reduction in American Cities. The uniform rate of consumption per meter and per metered consumer (estimated) as compared with the extremely variable consumption per unmetered consumer (estimated) indicates decisively a considerable and very variable amount of controllable wastage by unmetered consumers. If data could be obtained reigarding the quantities of water used for public and manufacturing purposes, as well as the number of manufacturing and domestic meters, it might be possible to indicate the probable amount of controllable wastage. Although about 37 per cent. of the services in Springfield are metered and house-to-house inspections are made over a considerable portion of the unmetered premises each year, the combined effect of these expedients is practically ineffective as far as controlling wastage is concerned. This is explained by the fact that metering is voluntary and at the expense of the consumers; hence only economical consumers, who probably would not in any event allow water to waste unnecessarily, apply for meters, and the resultant saving of water is on use, not on waste. House-to-house inspections have here been demonstrated as of practically no permanent value in the suppression of wastage. Their effect is only on a very small part of, the supply at one time and practically as soon as an inspector leaves the premises carelessness again prevails and wastage is resumed. SYRACUSE, N. Y. YRACUSE has had an interesting experience with the waste reduction problem and apparently has not yet secured the full results probably attainable, notwithstanding the fact that about half of the total consumption and 72 per cent. of the services are metered. Were it not that in recent years an attempt has been made to classify the amounts of water used for different purposes it would not be possible to explain the high rate of consumption or locate the principal sources of wastage. The following table gives the data regarding population, water consumption, consumption per capita per diem, number of services and meters in use, percentages of services metered, number of inhabitants per service and average daily consumption per service: 199 Waste of Water and Its Reduction. TABLE NO. 47. Average Daily Per cent. Average Average Con- No. of No. of of Servi- Inhab- Daily YEAR. Popula- Daily Con- sump- Services Me- ces to itants Contion. sump- tion in Use. ters. Me- Per sumption. Per ters. Service. tion Per Capita. Service. 1894 97,000 6,560,000 68.0 2,974 1,246 44 32.4 2,210 1895 99,000 7,000,000* 71.0 4,045 2,167 53 24.5 1,730 1896 101,000 7,740,000 77.0 7,436 3,696 50 13.6 1,040 1897 103,000 7,740,750 75.0 9,944 4,700* 47 10.3 780 1898 105,000 8,553,000 81.0 13,237 5,806 44 7.9 647 1899 107,000 10,030,000 93.0 13,500* 6,700* 50 7.9 744 1900 108,374 10,995,000 102.0 13,750* 8,037 58 7.9 800 1901 111,000 11,604,000 105.0 14,100* 8,600* 61 7.9 820 1902 113,000 11,927,000 105.0 14,598 9,247 63 7.7 816 1903 115,000 12,254,000 106.6 15,784 10,063 67 7.3 776 1904 117,000 11,962,000 102.0 15,969 11,059 72 7.3 747 *Approximate. Several interesting points are worthy of attention in this tabulation and are explained by the history of the plant. Prior to January 1st, 1892, the supply of water was under private control, the water having been supplied largely from springs and having been very limited in quantity; so much so, in fact, that the works were taxed to the limit to supply 2,700 water takers through about 40 miles of street mains. When the city purchased the works in 1892 steps were immediately taken to secure a new supply from Skaneateles Lake, and by July, 1894, the works were completed and the old supply discontinued. It will be seen that in 1894 there were but 2,974 services in use in a city with 97,000 inhabitants, a ratio of 32.4 persons per service. The per capita consumption given for that year, 68 gallons per day, and that for each of the four successive years, bears no relation whatever to the actual consumption per capita among the consumers and points out one of the dangers in making comparisons between the water consumption of different cities when all the conditions are not fully understood. Actually, during these early years the consumption per consumer was greatly in excess of that at the present day, as can be seen by a comparison of the relative amounts of water consumed per service. This great reduction in consumption per service, however, is not in any 200 Appendix D: Waste Reduction in American Cities. great measure due to suppression of wastage, but to the relatively large amounts of water used by trades and commercial interests and taken for public buildings, watering troughs, fountains, street sprinkling, etc., which amounts were nearly as great in 1894 as in 1904. An analysis of the distribution of the water used for different purposes in 1903 and 1904 is instructive in showing that at the present time probably almost as large a reduction of wastage can be made in water used for public purposes as in that taken for domestic consumption. In 1904 the distribution was as follows, in gallons per day: TABLE NO. 48. Metered Unmetered Water. Water. Total. Domestic consumption. 1,112,000 4,830,000* 5,942,000 Public uses........... 551,000 1,563,000 2,114,000 Commercial and miscellaneous.......... 4,606,000........ 4,606,000 Totals.......... 6,269,000 6,393,000 12,662,000 *Unmetered domestic consumption includes all unaccounted for water. Reducing these to gallons per capita per day, the for different purposes would be as follows: consumption Metered domestic consumption................. 9.5 Unmetered domestic consumption, including all waste..................................... 41.5 Total domestic consumption and unaccounted-for w ater.................................... Public uses metered............................ 4.7 Public uses not metered........................ 13.3 Total public uses............................... Commercial and manufacturing uses (metered). gals. per day 51.0 " " " 18.0 " " " 39.3 " " " 108.3 " " " Total per capita consumption............... As under very favorable conditions it would be unlikely that more than 80 per cent. of the total consumption could be accounted for, due to the under registration of meters, leaks from mains, 201 Waste of Water and Its Reduction. service pipes, etc., if 20 gallons be deducted from the unmetered domestic supply (which includes wastage and unaccounted for water), the probable per capita consumption would be at present as follows: Domestic consumption.................. 31.0 gals. per day Commercial and manufacturing uses..... 39.3 " i Public uses............................. 18.0 " " " Unaccounted for water.................. 20.0 " " " Total.............................. 108.3 " " " It will be noted that the total per capita consumption here reached is 108.3 gallons per day, as compared with 102 gallons given in the first table. The difference arises from the fact that in 1904 pitometer tests were made of the discharging capacity of the 19 miles of 30-inch conduit bringing the water to the city, and the result showed a discharge at the rate of about 700,000 gallons per day in excess of the calculated discharge arrived at by the methods employed in former years. The correction has been made in the foregoing calculations of per capita consumption by adding 700,000 gallons per day to the unmetered domestic consumption; but the quantities given in the first table have not been corrected, owing to uncertainty as to whether or not the same ratio of correction should be applied to the consumption in former years. During 1903 meters were placed on all public schools, nearly all the Fire Department buildings, the City Hall and the Public Library. In buildings where motors were operated by water a separate meter was placed on such service. The schools were metered in February. Table 49, on opposite page, shows the per capita consumption in gallons. It will be seen that with the exception of the Commercial, May and Willard schools, where excessive wastage had prevailed, the amounts used at the other schools had not been materially changed, indicating, as is usual, that wastage is confined to a relatively small number of consumers. No water for motor uses is included in the amounts given for the different schools. 202 Appendix D: Waste Reduction in American Cities. TABLE NO. 49. PER CAPITA CONSUMPTION IN GALLONS. Average SCHOOL.Daily SCHOOL. Attendance. Average for Average for Averafor (1903.) March, April October and the Year. and May. November. (1904.) (1903.) (1903.) Jefferson............... 314 21.0 16.9 31.6 Salina................. 612 0.1 0.1 0.1 Grant.................. 304 23.9 7.2 10.4 Townsend......... 481 5.5 5.6 7.8 Garfield................ 434 5.6 3.8 5.9 Franklin............... 826 3.3 7.0 9.8 Prescott................ 574 8.2 5.8 6.8 Clinton................ 522 15.4 17.3 23.6 Lincoln................ 414 8.3 8.4 12.2 Vine................... 218 3.5 4.5 5.7 Frazer................. 428 13.0 9.6 16.9 Genesee.............. 412 4.0 3.8 8.4 Commercial............. 86 425.0 70.5 108.1 May................... 426 114.0 36.2 64.6 Tompkins.............. 388 6.8 4.4 6.7 Porter.............. 778 4.7 2,9 5.0 Gere................... 352 29.7 27.5 35.5 Madison................ 529 37.6 29.7 36.5 Sumner................ 358 18.9 17.3 27.6 Washington Irving...... 439 29.6 27.1 23.9 Willard................ 81 161.4 75.5 66.6 Montgomery............ 471 4.4 4.4 5.6 Putnam................ 650 17.0 17.2 28.8 Andrew Jackson........ 353 3.4 3.7 6.4 Croton.................. 605 34.6 27.5 22.7 New High School...... 1,267 9.1 10.1 14.7 Seymour............... 665 24.5 24.3 50.0 Truant School.......... 16 36.7 12.2 33.0 Delaware............... 523 7.8 9.5 12.2 Grace.................. 259 24.0 27.4 25.4 Merrick................. 294 25.3 10.2 11.2 Bellevue............... 257 34.3 8.0 10.7 Danforth............... 348 23.1 18.4 17.3 Elmwood............... 224 12.5 43.9 43.9 Brighton............... 868 4.1 1.5 4.3 203 Waste of Water and Its Reduction. In percentage of the total supply the amounts of water used for different purposes in 1904 were as follows: METERED. per cent. Commercial uses............................ 26.0 Schools, Fire Dept. buildings, etc........... 4.4 Elevators and organ motors................. 10.4 Metered domestic consumption.............. 8.7 Total metered water.......................... 49.5 per cent. UN METERED. Unmetered (estimated uses)Flushing and sprinkling streets, bridges, etc. 4.1 Watering troughs and fountains............. 8.3 Unmetered domestic consumption, waste and unaccounted for water................... 38.1 Total unmetered water........................ 50.5 per cent. Grand total (metered and unmetered)............ 100.0 " " From this it will be seen that the water used for elevators, motors, watering troughs and fountains amounted to nearly onefifth of the entire supply. In the annual report for 1904 it is estimated that the average daily metered use for each domestic consumer is 81.2 gallons; and based on this it is stated that the 10,653 schedule consumers should ordinarily use 865,000 gallons of water per day, from which calculation it is deduced that a fair presumption would be that nearly 4,000,000 gallons daily is lost through waste in schedule services, underground leakage, under-registration of meters, under-estimating of unmetered public uses, fires, etc. Upon this assumption the total average daily domestic consumption would have been in 1904: Gallons Metered domestic consumption....................... 1,112,290 Schedule domestic consumers......................... 865,000 Total....................................... 1,977,290 which would have been equivalent to about 16 3-4 gallons per capita per diem based on the entire population of the city. SUMMARY.-No specific data in regard to waste reduction can yet be derived from the Syracuse records, owing to the rapid rate at which new services have been connected with the mains since the Skaneateles supply was introduced, and the large population for the relatively small number of consumers for the first four years thereafter. 204 Appendix D: Waste Reduction in American Cities. Practically half the entire supply is sold by measure and about 72 per cent. of the services are metered, yet the per capita consumption with about 7.3 persons per service is over 108 gallons per day. An unusually large proportion of the supply is used for public purposes, notably for watering troughs and fountains, and a relatively large amount for elevators and motors. The meters are placed and kept in repair by the city at the expense of the consumers. The use of meters is compulsory on the services of saloons, restaurants, drug stores, bakeries, dental offices with fountain jets, photograph galleries, printing offices, laundries, breweries, soda water manufacturers, factories, foundries, fountains, livery stables, steam engines, motors, greenhouses, business blocks, elevators, fire services, barber shops, urinals, machine shops and air pumps. The city can compel the use of meters on wasteful domestic consumers. The minimum charge by meter is $5.00 per year for domestic consumers, 36,000 cubic feet being allowed per family, with an additional charge for extra quantities according to the extra amount used. The meter rates are as follows: TABLE NO. 50. MINIMUM CHARGE FOR ANY SERVICE IS $5.00 PER ANNUM. MONTHLY METER RATES (IN CUBIC FEET): 1.900 or less............. $0.14 per 100 cu. ft. 1,900 to 2,400 inc., lump sum,..................... $2.65 2,400 to 6,000............ 0.11 6,000 to 8,300 inc., lump sum,..................... 6.60 8,300 to 14,400.............. 0.08 14,400 to 17,700 inc., lump sum.................... 11.50 17,700 to 21,000................ 0.06 21,000 to 27,300 inc., lump sum...................... 13.65 27,300 to 40,000............... 0.05 40,000 to 57,000 inc., lump sum....................... 20.00 57,100 and up................ 0.03~ 205 Waste of Water and Its Reduction. TAUNTON, MASS. TAUNTON is a city where, since 1898, all manufacturing and from 35 to 40 per cent. of the domestic consumers have been metered. The following tables relating to the water consumption, meters, services, and the proportion of the consumption used for different purposes, have been compiled from the Annual Reports of the Water Board: TABLE NO. 51. AVERAGE DAILY CON- NUMBER OF SUMPTION. Average PercentNo. of Daily Con- age of YEAR Con- sumption, Meters Meters to sumers. Gallons. Per Per Servi- _ Services. Cap- Con- ces. Domes- Manuita. sumer. tic. factur- Total.._________ ing. 1898 26,840 1,303,610 48 48.6 4,233 1,469 154 1,623 38.3 1899 26,890 1,443,177 53 53.7 4,372 1,572 161 1,733 39.6 1900 26,116 1,619,985 53 62.0 4,502 1,669 163 1,832 40.7 1901 28,000 1,729,354 56 65.0 4,618 1,769 165 1,934 41.8 1902 27,000 1,534,740 49 54.0 4,698 1,871 179 2,050 43.1 1903 27,350 1,519,629 49 54.0 4,753 1,933 201 2,134 44.8 1904 27,500 1,755,201 56 64.0 4,873........... 2,199 45.1 Average number of consumers per service, 1898-1904=5.93. TABLE NO. 52. PERCENTAGE OF TOTAL SUPPLY PASSED THROUGH METERS...- a- Unmetered Water and YEAR. Water Unaccounted For Domestic For Manufactur- Total. for. Consumption. ing Consumption. 1898........... 13.6 26.6 40.2 59.8 1899.......... 15.7 24.2 39.9 60.1 1900........... 13.7 23.8 37.5 62.5 1901........... 13.7 22.4 37.1 62.9 1902........... 15.7 28.2 43.9 56.1 1903........... 16.5 30.8 47.3 52.7 1904.................. 37.0 63.0 206 Appendix D: Waste Reduction in American Cities. While no condcusive deductions can be drawn from these records, it will be seen that although all manufacturing consumers are metered, there is an apparently irrational fluctuation in the consumption, chargeable, probably, to wastage from the fixtures of unmetered consumers. An indication that this is the case can be deduced from the records of metered domestic consumption and numbers of metered domestic services. If, for instance, it may be assumed that the consumption from the unmetered domestic services, if metered, would have been at the same rate as has obtained each year for the metered domestic consumers, the total legitimate domestic consumption each year would have been as follows: 1898.......18.3 gallons 1902.......20.5 gallons 1899.......22.5 (" 1903.......21.0 " 1900.......20.0 " 1904.......21.5 " 1901.......22.5 " which would have given for ideal conditions in Taunton the following distribution of the total consumption per consumer: TABLE NO. 53. WATER USED, IN GALLONS, PER CONSUMER PER DAY.. For Manufac- For Domestic YEAR. turing Purposes. Consumption. Total _________ Balance. Actual If Completely Consumption. Metered. MeIf tered. 1898........... 13. 18.3 17.3 48.6 1899........... 13. 22.5 18.2 53.7 1900........... 14.8 20.5 26.7 62.0 1901........... 14.6 22.5 27.9 65.0 1902........... 15.2 20.5 18.3 54.0 1903........... 16.6 21.0 16.4 54.0 1904........... 14.7* 21.5 27.8 64.0 *Estimated. 207 Waste of Water and Its Reduction. TABLE NO. 54. DISTRIBUTION OF CONSUMPTION IN GALLONS PER CONSUMER PER DIEM. MANUFAC- DOMESTIC CONSUMERS, PUBLIC TURING USES, AND UNACCOUNTED-FOR YEAR. CONSUMERS. WATER. Total. All Metered. Metered.** Unmetered. 1898........... 13.0 6.6 29.0 48.6 1899........... 13.0 8.4 32.3 53.7 1900........... 14.8 8.5 38.7 62.0 1901.......... 14.6 8.9 41.5 65.0 1902........... 15.2 8.5 30.3 54.0 1903........... 16.6 8.9 28.5 54.0 1904........... 14.7* 9.0 40.3 64.0 *Estimated. **Part of Domestic. While the above figures for domestic consumption are estimates, based on the rate of consumption of the metered domestic consumers, and therefore not claimed to be exact, it is probable that the error in each case is not great enough to affect the general conclusion that the fluctuations in the amounts of water in the column headed "Balance" (which are the differences between the actual per capita consumption per consumer and the sums of the actual manufacturing and estimated domestic consumptions) are indications of considerable preventable wastage controllable by complete metering. While sufficient data are not available to carry this analysis further, it is significant that with all the manufacturing and nearly half the domestic consumption metered, the water not accounted for (assuming the other half of the domestic consumers to be metered, and consuming water at the same rate as those now metered) should still have amounted to 43 per cent. of the entire consumption in 1904. The meters are the property of the consumers, but are set by the Water Department. In addition to testing meters when suspected of incorrect registration, the Department makes a systematic practice of testing all meters when they shall have been in use for ten years. As a summary of many years' experience with meters in Taunton, the 208 Appendix D: Waste Reduction in American Cities. Water Board concludes that the leading manufacturers of water meters make devices which, on a general average, are durable, accurate and trustworthy, but that there are individual faulty meters of all makes. In over 13,000 tests the average recorded quantities of water passed were slightly less than the actual quantities, indicating, on the whole, that when the meters are properly cared for, they are fair and equitable in their records both for the supplier and the consumer, with a general leaning in favor of the latter. SUMMARY.-With all manufacturing and about 45 per cent. of the domestic consumption in Taunton metered there is still a considerable wastage of water, the exact amount of which cannot be determined from the records; but its extent in recent years has been quite variable, and probably rarely less than 20 per cent. of the entire supply, estimated allowances being made for water used for public purposes, under-registration of meters, slippage of pumps, leakage from mains, etc. The most important deduction seems to be that, as in most cities, the stoppage of wastage on half the domestic consumers is not sufficient to give satisfactory relief, since the other half may in certain years waste as much as all the domestic consumers combined have wasted in other years. WELLESLEY, MASSACHUSETTS. W ELLESLEY is a small residential city in which at the present time all water is sold by meter. The entire pumpage, with the exception of that used for extinguishing fires and for flushing out dead ends, passes to consumers through meters. In 1888 the pumpage had increased so very materially, with no proportional increase in population, that excessive wastage was apparent, particularly in very cold and very hot weather. In 1892 the Water Commissioners decided that the only way to reduce the excessive consumption was to make a general application of meters and a resolution was therefore adopted to meter all consumers having more than one fixture and charge a minimum price of $10.00 per year and a rental of $2.00 per year and upward for the meters, according to size. This rate entitled the consumer to 33,333 gallons per year, any excess above this quantity to be charged for at 25 cents per 1,000 gallons. 209 Waste of Water and Its Reduction. By the end of 1893 practically all consumers were metered and the consumption had been reduced over 30 per cent. During that year the meter rates were readjusted to 25 cents per 1,000 gallons with a minimum charge of $6.00 per annum. In 1894 the town acquired by purchase all the private meters then in use. During 1897 tests were made to locate leaks in the mains and service pipes, resulting in the discovery of several small leaks but none of consequence. Prior to 1900 the meter rates were altered to allow each taker 16,000 gallons per annum for the minimum price of $6.00 in advance and 25 cents per 1,000 gallons for any excess, with no rental charge for the use of the meters. These rates are now in force. The following table summarizes the statistics regarding the consumption of water and use of meters in Wellesley: TABLE NO. 55. AVERAGE DAILY CON- PROPORTION OP SUMPTION, IN NUMBER CONSUMERS GALLONS. OF METERED. Pop- Con- ____ YEAR. ula- sumtion. ers. Per Per DoIn- Con- Ser- Me- mes- ManufacTotal. hab- sum- vices. ters. tic. turing. itant. er. No 1887....... 3,300 1,500 105,490 32 70 302 48 data No data 1888....... 3,400 1,726 138,156 40 80 412 66 " " 1889*..................................... 1890....... 3,590 2,650 255,418 71 96 503 93 1891*............................ I. 1892........ 3,625 2,866 239,520 66 84 84 864 1 1893...... 3,800 2,901 200,871 55 71 610 566 All Pract'llyall 1894..... 4,000 2,933 186,037 46 63 647 588.... 1895..... 4,229 3,030 175,345 41 58 685 656.... 1896.................................. 1897....... 4,500 3,196 176,714 39 55 766 16... 1898....... 4,650 3,874 166,765 36 44 795 737 1899*..................... 1900....... 5,072 4,929 238,884 47 48 878 799...... 1901....... 5,240 5,097 245,220 47 48 898 854 " All 1902....... 5,350 5,147 257,001 48 50 893 877 " 1903....... 5,417 5,324 293,704 54 55 916 911 " 1904....... 5,793 5,700 314,219 54 55 949 953 " *For the years omitted it has been impossible to secure copies of the Annual Reports of the Water Board. 210 Appendix D: Waste Reduction in American Cities. We have here a carefully kept record of the consumption of water in a strictly residential town where, for the last five years, every consumer has been supplied through a meter and where all water used can be separately accounted for except that used for extinguishing fires and flushing out the mains. All the water is pumped and therefore the unaccounted-for water would include the slippage of the pump plungers, the under-registration of meters, the water used for fires and for flushing street mains, and the leakage from mains and service pipes outside the houses. The detailed distribution of the water for the years 1901-1904 inclusive was as follows: TABLE NO. 56. WATER USED BY Unaccounted Used for for Water incStreet cluding Slip of Domestic Manufac- Sprinklin Pumps and TotalConYEAR. Cnsumers. turing Meters, Wa- Sumption Per Consumers. ter for Fires, Day,.,Per Flushing of Consumer. Mains and Metered. Metered. Estimated. Leakage. 1901........ 24.3 0.3 2.8 20.6 48.0 1902........ 25.4 0.3 2.2 22.1 50.0 1903........ 28.2 0.4 3.2 23.2 55.0 1904........ 28.6 0.4 2.5 23.5 55.0 The unaccounted-for water, therefore, averages for the four years 43 per cent of the reported pumpage. An allowance of 5 per cent. for under-registration of meters and 15 per cent. for the slippage of the pumps would make the average daily domestic consumption for these four years 28 gallons per capita and the total consumption 45, leaving 14 gallons or 31 per cent. of the total to cover water used for fires, flushing mains and water possibly leaking from mains and service pipes. On examining the consumption of water per consumer and per inhabitant, it will be noted that in 1887 the apparent consumption was only 32 gallons per inhabitant, whereas the actual consumption per consumer was 70 gallons per day. With no means in force to restrict wastage the consumption rapidly increased until 1890 when it reached 96 gallons per consumer. During the 211 Waste of Water and Its Reduction. next three years meters were placed on all domestic, and practically all manufacturing consumers, with the result that by 1895 wastage had been reduced by 38 gallons per consumer per day, representing a saving of 40 per cent. of the total pumpage. Since 1895 the amount of water pumped per capita has averaged for each year considerably less than this, the least yearly average having been 43 gallons in 1898, which was less than that in 1890 by 48 gallons per day. If the pumpage were metered instead of calculated from plunger displacements, the total consumption would be less than given in the tabulations and the domestic consumption somewhat more, thus reducing the amount of unaccounted for water. Probably the average actual consumption per capita among the domestic consumers is about 30 gallons per capita daily, the manufacturing consumption only about 0.5 gallons per capita, and the unaccounted for water not over 14 gallons per capita. As a rule there is very little known regarding the slippage of pumps in water works stations. Most Superintendents allow from 4 to 10 per cent., but it is probable that there are few pumps in operation which, in a yearly average, will check up within 10 per cent. of the true quantity of water pumped. SUMMARY.-The reduction of the consumption of water in an amount equal to 30, and possibly to 40, per cent. of the pumpage has been effected by metering every consumer in the city; and this reduction represents the suppression of useless wastage on the part of domestic consumers only. The results have been satisfactory and permanent. At the present time the domestic consumption in this strictly resdential city is apparently about 30 gallons per day per consumer, and the unaccounted for water an unknown amount, owing to uncertainty in the estimation of the total pumpage. WEST ORANGE, NEW JERSEY. WD EST ORANGE, a suburban residential village with about 6,000 inhabitants, is supplied with water by a private company which purchases its water from the Montclair Water 212 Appendix D: Waste Reduction in American Cities. Company by meter measurement and sells it by meter to all consumers. Being a private corporation, no details or statistical data have been published. Regarding the accounting for the water sold, Mr. Carroll Ph. Basse.i, C. E., Ph. D., Consulting Engineer, writes: "In West Orange we buy our water by meter and sell it largely by meter, and we have found it practically impossible to account for more than 80 per cent. of the amount received on a system of 30 miles of pipe, with a pressure varying from 40 pounds to 115 pounds; but we keep reasonably close to this percentage most of the time by great care in shutting out leaks." YONKERS, NEW YORK. HE water supply of Yonkers is pumped and sold entirely by meter. In the following Tables no allowance has been made for the slippage of the pumps nor for the under-registration of meters, no reliable data having been secured by which such allowances could accurately be made. The records are therefore of interest in demonstrating that where even an approximate estimate can be made of the quantity of water furnished by computation from pump plunger displacements, and when all water is sold and used through meters, it is still impossible to account for nearly half of the total quantity supposed to have been supplied. In this respect the records of these Water Works are invaluable. In many cities arbitrary corrections are made for these losses, with no statements of the basis upon which they were made, and hence from such records it is impossible to deduce exact data. In the case of Yonkers it is different. No allowances have been made and the quantities given show precisely the conditions that have obtained. The following table shows the population, average daily consumption per consumer, number of services and number of meters in use: 213 Waste of Water and Its Reduction. TABLE NO. 57. Average Popula- Average Daily No. of No. of Miles No. of YEAR. tion Daily Con- Consump- Servi- Me- of ConsumSupplied. sumption, tion Per ces. ters. Mains. ers Per Gallons.* Consumer, Service. Gallons. 1900....... 46,500 3,626,564 78 4,968 4,852 80.08 9.4 1901.. '... 48,000 4,098,509 84 5,238 5,012 84.66 9.3 1902....... 51,000 4,540,924 89 5,495 5,249 88.84 9.3 1903....... 55,000 4,777,452 88 5,675 5,644 91.19 9.7 1904....... 59,000 5,529,531 94 5,918 5,981 92.99 9.9 *No allowance made for slip of pumps. TABLE NO. 58. DISTRIBUTION OF SUPPLY IN GALLONS PER CAPITA PER DAY. Manu- Public Special UnacYEAR. factur- Domestic. Uses. Uses.** counted Total. ing. For.* 1900.......... 13 18.4 9.0 37.6 78 1901.......... 16 19 3.0 4.6 42.4 84 1902.......... 22 19 3.0 5.0 40.0 89 1903.......... 22 20 2.0 7.0 37.0 88 1904.......... 24 20 2.0 7.5 40.5 94 *Includes slippage of pumps, under-registration of meters and all waste and other losses. **Mainly water furnished to 24th Ward of N. Y. City. Practically all the water drawn from the mains in Yonkers for domestic, manufacturing and public uses is metered and hence can be accounted for as fully as in any American city where the total supply is not measured as it leaves the pumping stations, the only uncertainties being in the amount of slip to allow for the pumps and the loss by under-registration of the meters. From the previous Table it will be seen that without making allowances for slip and under-registration the total amount of unaccounted fpr water has varied from 37 to 42 gallons per consumer per day. SuMMARY.-All the water in Yonkers is sold by measure. The reported pumpage is not corrected for slip and hence is too high. No figures are given by which to apply corrections in order to get at the correct amounts. Without making any correc214 Appendix D: Waste Reduction in American Cities. tions the domestic consumption is about 20 gallons per capita per day per consumer, and the unaccounted for water about twice that amount. Actually, judging from the records of cities receiving their whole supply through meters, and selling through meters, it is probably not likely that over 80 per cent. of the total supply could be accounted for in any event. If this were the case in Yonkers, the accounted for water in 1904, say 54 gallons per capita, would represent 80 per cent. of 68 gallons, or 28 gallons, deducting which from 94, the reported pumpage would leave 26 gallons per capita per day to represent the slippage of the pumps, or, about 28~2 per cent. This may seem a high figure to the general Water Works Superintendent; but recent tests in Philadelphia of the very large well-built, high duty pumping engines, showed a slippage of from 18 to 56 per cent. in the individual engines and an average of 25 per cent. for all the pumping stations in the City. No data for costs of meters or meter maintenance are available. CHICAGO, ILLINOIS. I N comparison with the foregoing data it may be well, for the.purpose of exhibiting the rapid increase in rate of consumption sure to follow where no effort is made to check wastage, to quote the records of the Water Department of Chicago, Ills. This City has been chosen for illustration because its records run back to the first settlement, in 1831, on the site of the present City, thus giving a complete history of the works; and for the further reason that at the present time, although a comparatively young City, the total amount of water supplied daily, according to the reported pumpage, is greater than any other city in the world, although not at as great a rate per capita as in at least two other American cities. It is to be understood that the pumpage reported is probably much in excess of the true amount actually delivered into the street mains, but no data are available for making the necessary corrections. Not much can be learned regarding waste reduction from the 215 Waste of Water and Its Reduction. Chicago records. No systematic attempts have been made to reduce wastage further than occasional inspections and a few observations on day and night flow in the mains in certain parts of the city. Chicago has increased in population at such a rapid rate that the history of the water works is a record of successive additions to the pumping capacity of the various plants furnishing the water. No city in the world can equal the record Chicago has made in this respect. All the water consumed in the city is drawn from Lake Michigan through tunnels reaching out from 2 to 4 miles under the lake (with the exception of a very insignificant amount of ground water) and pumped into the mains in the streets, there being no reservoirs to serve as compensating basins and give a reserve supply. No city in the world is, or ever has been supplied with so great a quantity of water per day as is being pumped into the distribution system in Chicago to-day. The average daily pumpage during 1904 was practically 400,000,000 gallons, to which figure the consumption has grown, from a few hundred thousand, in about 55 years. It can be appreciated that the problems that have been presented to the managers of this stupendous organization have been unique and without precedent. An examination of Table 59 opposite, which contains the population, average daily consumption of water, average daily consumption per capita, number of services in use, number of meters in use, average daily consumption per service, and data regarding the successive additions to the pumping plants will prove of interest. A comparison of the per capita consumption with the consumption per service shows that in the early days but a small proportion of the population was supplied with water. The per capita consumption in 1858 was very low, yet the consumption per service was fairly high and has since then increased but about 50 per cent.; while during the same length of time the per capita consumption has increased five fold. The available data are too meager to analyze the significance of the increase in consumption per service. During the past ten years the increase was at a greater rate than during the ten year period immediately preceding; and this period, in turn, showed a greater increase than the one before. The increase in the number of meters installed has also been at a greater rate during recent years. 216 Appendix D: Waste Reduction in American Cities. TABLE NO. 59. Average =: 4 Popula- Daily Con. ofNo f No. of Increasd in PumpYEAR. tion. Sumption, CO~ Servi- Me- ' ing Capacity. Gallons. ~, ces. ters. 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 60 600 350 1,800 3,265 4,000 4,179 4,000 4,200 4,470 5,500 6,590 7,580 8,000 12,088 14,169 16,859 20,023 23,047 28,296 24,000 38,734 60,662 65,872 80,023 86,000 93,000 84,000 94,000 109,260 120,000 137,030 150,000 161,288 178,492 200,418 225,000 252,054 279,329 306,605 337,000 367,396 381,402 395,408 401,534 407,661 422,731 436,731 464,123 491,516 526,104 560,693 595,339 629,985 666,901 703,817 753,234 802,651 1,005,660 1,208,669 1,323,339 1,438,010 1,502,868 1,567,727 1,592,183 1,616,635 1,734,111 1,851,588 1,929,641 2,007,695 1,786,226 1,844,661 1,903,096 1,962,251........ 2,392,945... 4,000,000... 3,552,052... 6,074,739 6,400,298 6,9132059 7,6910,459 11,562,273 14,724,999 281,766260 23,464,287 27,56,81953 32115627,31 38,090,952 39,844,556 41,931,481 523,183,892 2753,60,7819 56322,44171 57384,37695 63,922,700 73,265,592 97,789,706 104,315,362 110,8952,707 174,114,254 236,43417699 251,7839,816 254,2078,52 26,35,462,53 273,055,816 300.614,394 3742,824,244 376,015,974 2398,952,350 29.9 47.5 38.2 35.6 41.2 43.0 40.3 44.3 42.6 42.8 42.6 43.3 51.3 58.4 66.7 70.9 69.6 74.9 84.2 96.3 99.2 102.8 123.6 122.7 121.3 116.7 121.5 118.0 123.0 127.0 137.4 138.9 135.3 129.9 110.2 126.0 131.5 135.0 157.3 152.1 158.1 157.3 153.0 147.4 155.7 160.6 191.9 194.1 196.0 203.3 6,350 6,876 7,450 8,412 9,855 11,406 13,634 16,498 20,915 28,308 35,318 38,110 42,511 47,995 51,563 55,293 57,130 59,369 61,384 63,510 67,949 73,627 78,840 85,496 92,133 98,688 106,771 115,578 125,667 "133,284 155,096 182,382 203,954 242,208 258,092 272,758 276,240 286,179 292,386 299,901 308,945 315,954 324,202 331,506 339,044 ~65 99 133 147 179 253 400 530 656 341 618 836 1,174 1,313 1,446 1,623 1,888 2,067 2,113 2,163 2,310 2,503 2,685 3,8085 3,273 3,122 3,287 3,924 4,131 4,326 4,654 4,878 5,120 5,106 5,396 15,837 6,8 6,319'6 6,728 7,075 7,298 8,102 706 740 705 815 760 701 657 637 640 703 658 616 616 648 670 740 720 734 879 874 887 845 867 838 857 869 930 925 880 830 832 980 955 952 975 923 923 921 927 935 1,022 1,044 1,087 1,105 1,136 1.177 T'n of Chicago incorp'ted City of Chicago incorp't'd First water pipes laid. Works acquired by City. 8,000,000 gallons. 13,000,000 18,000,000 36,000,000 30,000,000 30,000,000 24,000,000 49,500,000 36,000,000 66,800,000 37,500,000 250,000 14,000,000 60,000,000 61,500,000 60,000,000 25.000.000 Nowe-Additional pumping engines under construction and to be in operation soein time between 1905 and 1909; Capacity, 153,350,000 gallons. 217 Waste of Water and Its Reduction. From 1891 to 1900 the receipts from the sale of water through meters, although only about 2>8 per cent. of the services were metered, averaged about half as much as the receipts from assessment rates, and furnished approximately one-third the total receipts of the Department. The Reports of the Department do not give separately the number of gallons metered and unmetered, nor do they state whether the metered water includes all the uses not classed as strictly domestic. In fact, from the distribution of meters under different classifications it appears that in 1902 over 1,200 of the 7,000 then in use were on the services of apartment houses and residences. In the absence of specific data it is impossible to frame any hypothesis from the circumstance that about 28 per cent. of the services deliver the water from which 33 per cent. of the revenue is derived, except that consumers supplied through the remaining 98 per cent. of unmetered services are paying for the water at only 1-25th the rate per 1,000 gallons paid by the 2 per cent. of metered customers if calculated on the quantity of water delivered into the mains. This, of course, is the apparent condition: The true condition is that if wastage were reduced to the least practicable amount the assessment rates would remain the same, but the quantity of water pumped would be reduced and extensions of the works would not have to be made at such short intervals; every extension, of course, being accompanied by an increase of the City's debt and consequently by increased taxation. The City authorities are, it is needless to say, awake to the necessities of the case, although until about 1900 no definite recommendations regarding waste reduction appear to have been seriously considered. In 1901-1902 Mr. John Ericson, City Engineer, presented a special report upon the subject to the Commissioner of Public Works in which he recommended metering all consumers and testing the mains. These recommendations were approved by the Commissioner and transmitted to the Mayor and Councils, but were not adopted. Mr. Ericson, in a special report dated May, 1905, upon the Present, Past and Future Water Supply of Chicago, addressed to the Commissioner of Public Works, again urged the adoption of meters as a remedial measure, summarizing his conclusion as follows: 218 Appendix D: Waste Reduction in American Cities. "First.-The phenomenal growth of the City, together with the fact that large territories with independent water systems were annexed from time to time, made it impossible both from an engineering and a financial point of view to develop an ideal system for the metropolis. "Second.-The Water Works System, at the present rate of consumption, is sufficient for less than 1,850,000, while the present population is about 2,000,000. "Third.-Improvements and additions now authorized and under way, and the cost of which will be about $4,370,000, will not be fully completed until the year 1909. The total nominal daily capacity of the system will then be at the rate of 687,000,000 gallons, which, allowing for proper reserve, will be barely sufficient for a population of about 2,400,000. "Fourth.-Further additions, estimated to cost about $5,080,000, and which should be commenced in about 1907, or before, will bring the total daily capacity to about 940,000,000 gallons, which, allowing for proper reserve, will be sufficient for at the very most 3,000,000 people. This, it is estimated, the City will reach in about the year 1919. Still further extensions, not referred to in this Report, should therefore be commenced not later than 1916. "Fifth.-There is a probable leakage and waste of water pumped, of about 75 per cent. "Sixth.-The metering of about 40 per cent. of the taps in use on the System would undoubtedly reduce this waste and leakage to a great extent and make it possible to extend the time of completion of the various additions to the Water Supply System several years, and would, besides, cause a great saving to the City." PART II: GERMAN CITIES. GROUP I: Cities in Which Waste Reduction Has Been Accomplished by Metering. BERLIN. ERLIN has always had what the United States would be called a low per capita consumption. The highest amount consumed per person in any year, for nearly four decades, was 27.5 gallons per day, in 1860. At the present time the rate is about 18 gallons per capita per day. The lowest figure touched was in 1890 when the average use per capita per day was 16.1 gallons. It would seem, to one used to the conditions in the United States, which are so very different from those in most foreign countries, particularly in Germany, that the highest of these figures would not imply excessive wastage, yet it was so considered in Berlin and active steps were taken to cut it down. The concession under which the Water Company operated was to expire in 1881 and negotiations were in progress some years earlier for the prolongation of the term of years named. The purchase of the works by the Municipality, however, in 1873, obviated this necessity. While negotiations were in progress the company was in a critical condition. The consumption had reached the limit for which the works were designed, and in 1865 extensions became necessary. The waste of water was excessive: nor could it be prevented or even checked by house inspections. The company had the power to cut off the supply, yet this, even in the case of gross, persistent abuse had to be exercised with great discretion. Under these conditions it was imperative to adopt a policy that would husband the supply as much as possible and yet not check the rapid increase in customers then taking place, and to satisfy all. At this time Mr. Henry Gill, who had been Engineer and Manager of the Works since their opening, recommended a tentative change in the method of charge for water, with a view to 220 Appendix D: Waste Reduction in German Cities. its eventual sale exclusively by measure. This was approved and the necessary capital furnished. Under the concession the charges for water prescribed a constant supply delivered under a pressure sufficient to reach the highest floors, at a maximum height of 66 feet above the street, and allowed a charge of 4 per cent. on the house rent for water for domestic purposes, including the supply for baths and waterclosets. Extra charges were permissible for trades and garden uses. The water was then drawn directly from the mains, there being, as a rule, no tanks, even for water-closets, within the houses. It was determined that the Company should own and install the meters and examine and change them whenever desirable. The old rates prescribed that water would only be delivered by meter when a fixed charge of about $20.00 per quarter was guaranteed by contract. For this sum the consumer was entitled to about 233,600 gallons; for each 264 gallons additional the charge was about 1.8 cents, or 7.1 cents per 1,000 gallons. This stipulation almost precluded the sale of water by meter for domestic purposes. The new meter rates were adjusted to a figure by which, with a very liberal allowance of water per head per quarter, the charges would be less than 1 per cent. per quarter on the rental. This was made the minimum rate; all water used above this quantity was charged for at the tariff rate. The offers of a supply on these terms were readily accepted by a class of con-:sumers chosen by the Company. This little body of meter services was closely watched, the owners being warned when the fixtures were found defective, and the meters occasionally changed and tested. When it became well established that the meters were reliable their introduction slowly advanced, the minimum rate being reduced occasionally to correspond with the diminutions of the quarterly minimum quantity of water. From 1870 the process was pushed forward as rapidly as possible, the company being satisfied with the results and the consumers pleased with the reduction in the quarterly rate. In May, 1878, the Municipality, at the recommendation of Mr. Gill, resolved to discontinue altogether the sale of water on 221 Waste of Water and Its Reduction. TABLE NO. 60. FILTERED WATER, PUMPED. NUMBER OF PERCENTAGE TAKEN BY Average Plant Number of Daily Con- UnmeYEAR. Consumers. sumption Con- tered Per Capita, Properties Meters sumers Public tred Gallons. Supplied. in Use. Through Uses. su Meters. ers, Waste. 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879-80 1880-81 1881-82 1882-83 1883-84 1884-85 1885-86 1886-87 1887-88 1888-89 1889-90 1890-91 1891-92 1892-93 1893-94 1894-95 1895-96 65,805 80,463 88,622 147,963 179,247 211,137 229,255 243,829 265,010 342,696 423,915 450,492 484,900 522,240 492,631 502,329 555,782 712,224 768,096 842,803 883,555 954,770 981,158 1,065,705 1,095,858 1,126,012 1,221,162 1,316,312 1,360,285 1,388,530 1,427,148 1,606,424 1,645,426 1,674,221 1,703,481 1,719,183 27.5 25.9 22.7 21.7 21.4 23.2 23.8 23.2 26.1 20.9 18.7 17.7 17.7 17.7 20.1 20.9 20.3 20.3 17.7 16.6 16.6 16.9 16.9 16.9 17.2 16.9 17.2 17.2 16.9 16.1 18.0 16.4 17.7 18.2 18.0 18.2 669 886 1,134 1,466 1,822 2,359 3,274 3,941 4,686 5,062 5,500 5,914 6,283 6,598 6,915 7,524 8,114 8,666 9,079 9,649 12,365 14,001 15,616 15,750 16,576 17,034 17,654 18,216 18,659 19,193 19,775 20,403 21,324 21,437 22,180 22,638 23,042 23,428 23,912 97 132 138 166 190 232 312 367 464 616 742 896 1,038 1,203 2,404 3,725 5,297 6,277 6,919 7,502 9,705 13,885 15,544 16,350 16,789 17,255 17,679 18,178 18,678 19,775 19,803 20,416 21,031 21,625 22,302 22,815 23,229 24,321 25,428 5 15 17 27 28 35 28 29 30 31 34 36 39 43 52 67 76 82 81 78 75 79 78 82 82 82 83 82 85 84 84 86 87 88 88 86 86 85 82 32 25 24 27 23 24 21 15 14 14 14 12 11 10 9 13 16 15 10 9 9 9 9 10 10 9 10 10 10 10 9 9 10 9 9 11 11 10 9 53 60 59 46 49 40 51 56 56 56 52 52 50 47 39 20 8 3 9 13 16 12 13 8 8 9 7 8 5 6 7 5 3 3 3 3 3 5 9 222 Appendix D: Waste Reduction in German Cities. the percentage of rental basis and from the first of October of,that year to sell water like gas, solely by measure. Table 60 on page 222 gives the number of consumers, consumption per capita per day, number of properties supplied with water, number of meters in use and percentage of the total supply taken by consumers through meters, as well as the percentage used for public purposes and that used at the plant, which latter includes waste, and to 1872, the unmetered consumer's water. An examination of the per capita column of this table does not show what, for American conditions, would be considered a very great total reduction, only about 8 or 9 gallons per capita per day, yet this small amount represented in Berlin the prevention of wastage equal to about one-third the domestic consumption of the City. The last column in the table shows that after meters had been placed on all domestic consumers the wastage had been reduced to such a degree that it was possible to account for all water pumped within 1 or 2 gallons per capita per day. It is doubtful if there are many communities in which it would be profitable to attempt to reduce so small a total amount of wastage. In Berlin the works are owned by the municipality, which also owns and controls its sewage disposal works; and as practically all the domestic water eventually becomes sewage an increase in water supply per capita means not only increased cost for water but increased taxes for purifying the sewage, which is done on an extensive and elaborate plan at Berlin. Also, the cost of installing, maintaining and reading meters, per capita, is exceeding low for the reason that Berlin is built up very compactly, almost exclusively on the apartment house plan, *each building being in reality several dwellings on one foundation and under one roof, and in almost every case one metered service pipe supplies all the dwellings in the building; thus, on looking over the number of services and the population in 1896, it will be seen that each service pipe supplied water to upwards of 70 people. From the records of this city since about 1870 the following deductions seem reasonable: 1. Partial metering checked wastage but did not permanently control it. 223 Waste of Water and Its Reduction. 2. Complete metering has been an effective and lasting prevention of waste. 3. There has been a legitimate increase in the use of water per capita with the increase of population. 4. It is possible in this German city to account for all but one or two gallons per capita per day of the water pumped. 5. It was possible in this city, where the majority of the people are, and always have been, renters, (the renters deriving no direct benefit from the economical use of water and therefore having no personal incentive to be saving in its use) to educate landlords to voluntarily put meters on services by making rates such that it appealed to the owners as being more economical to buy their water by measure than to pay for it a percentage of the rental values of the properties. Many years were required to bring about this result, which would have been impossible but for the wise policy of the company, under which, through a nucleus of selected consumers, it became more and more widely known that metering was economical for the consumer, the landlord, and the city. It was also learned that practically all the water wasted ran from leaky faucets, closet valves and rusted-out service pipes. That when these were put in decent order, wastage, that is, noticeable wastage, ceased. That tenants were not wilfully wasteful, and that the landlords, therefore, ran no material risk in paying for water by measure when the rents included the furnishing of water to the tenants. That practically all the wasted water, or at least a very great percentage of it, was chargeable directly to the carelessness of landlords and owners rather than to the wastefulness of tenants. 224 .2" ,4-i, diI TABLE NO. 61. GROUP I.-Continued. CITIES IN WHICH WASTE REDUCTION HAS BEEN ACCOMPLISHED BY METERING. - -----------— I I --- Average Percentage of Daily Number of Supply CITY AND YEAR' Popu- Con- metered lation sumption Per Capita, Servi- Meters Con- Total Gallons ces sumers BAMr~BERG: 1888.ME 32,000 28.5 748 113 39 25 1889......... 33,000 26.9 842 113 36 23 1890.............. 35,000........ 113 34 22 1891............... 35,815 18.2 1,040 113 31 23 1892............... 35,980 19.2 1,091 138 34 25 1893............. 36,000 19.5 1,133 303 41 27 1894.............. 38,000 19.0 1,197 454 66 37 1895............... 38,949 15.3 1,247 506 71 44 1896..............30,000 15.3 1,318 608 73 44 1897............... 39,500 15.8 1,347 768 78 47 EISENACH: 1888...............20,000 33.0........ 1 26 18 1889............. 20,000 19.8........ 35 24 1890...............21,000 18.5....... 33 23 1891..............21,539 21.7....... 35 25 1892............... 22,094 16.6....... 67 43 1893.............. 22,500 15.8........ 93 54 1894............... 24,868 12.4........ 99 49 1895............... 26,347 14.3........ 99 45 1896............ 27,168 15.2....... 99 54 1897..............28,735 14.5........ 99 56 1898.............. 30,723 13.2........ 99 61 1899.............. 33,888 14.3........ 99 67 HAMBURG: 1858............... 198,350 24.8................ 1859...............206,800 25.3................ 1860..............211,500 26.1................ 1861............216,200 26.4................ 1862.............220,900 26.7................ 1863.............225,600 27.2................ 1864..............230,300 29.3................ 1865...............235,000 31.6.............. 1866..............245,000 35.0................ 1867.........2....256,000 37.2................ 1868...............266,000 38.8................ 1869...............277,300 41.4................ 1870............280,000 41.2................ 1871............... 287,300 44.9................ 1872...............296,000 45.6................ 1873.............. 304,000 43.6................ 1874.............I 317,000 45.2................ 1875............... 337,600 45.4................ 1876............... 341,700 47.4................ 1877...............353,300 48.8................ 1878............... 363,700 51.0............ 1879...............373,700 55.0................ 1880............... 396,900 50.0................ 1881............ 405,900 52.4................ 1882...............418,400 54.6............... 1883............... 431,100 59.6................ 1884.............447.400 63.3................ 1885............... 467,300 53.8................ 1886...............477,900 53.6................ 1887...............492.500 54.0................. 1888.........3..516,100 55.0............... 1889...............543,200 55.6 1890.........:.. 568,700 54.0 18,100 4,814 12(b) 27 1891............. 3 583,700 58.3 18,250 5,325 13(b) 29 1892...............580,800 58.0 18,400 5,776 14(b) 31 1893...............593,400 53.6 18,343 6,342 16(b) 34 1894............. 604,700 51.6 18,800 6,763 16(b) 36 1895..............620,400 52.2 19,049 7,787 17(b) 41 1896..............636,200 49.6 19,301 8,485 18(b) 44 1897............654,800 49.8 19,522 8,755 19(b) 45 1898.............668,700 48.6 19,823 8,974 20(b) 46 1899...............684.400 47.0 20.036 9,888 21(b) 49 HANOVER: 1888-1889..........105,800 32.8 5,832 288 18 18 1889-1890..........178,000 34.0 6,192 341 20 19 1890-1891..........189,976 31.2 7,132 415 17 17 1891-1892..........189,976 29.0 7,508 3,295 50 23 1892-1893..........189,976 25.6 7,411 7,566 100 60 1893-1894..........220,800 15.8 8,057 8,057 100 68 1894-1895........231,800 24.3 8,517 8,517 100 72 1895-1896..........246,000 21.2 8,894 8,894 100 76 Percentage Taken By Con- Pob- Plant c lic and sumer ^ ^ somers Uses Wast 64 33 3 63 34 3 65 32 3 73 24 3 73 24 3 67 25 8 57 35 8 63 24 13 60 24 16 62 21 17 69 24 7 69 28 3 71 27 2 70 27 3 64 35 1 57 38 5 49 46 5 46 43 11 55 35 10 57 34 9 62 28 10 67 24 9 Li It te, REMARKS 96 95 95 93 93 94 94 94 94 94 97.8 97.5 96.6 93.8 93.5 68.5 69.6 74.3 LEIPZIG: 1875............... 1876............... 1877............... 1878................ 1879............... 1880............... 1881............... 1882............... 1883............... 1884............... 1885............... 1886............... 1887............... 1888............... 1889............... 1890............... 1891............... 1892............... 1893............... 1894............... 1895............... 1896............ 1897............... 1898............... OFFENBACH: 1892-1893.......... 1893-1894.......... 1894-1895.......... 1895-1896.......... 1896-1897....... 1897-1898.......... 1898-1899.......... 1899-1900.......... RUDOLSTADT: 1888............... 1889............... 1890............. 1891............. 1892............... 1893............... 1894............... 185............... 1896............... 1897.............. 1898.............. 1899............... STETTIN: 1888-1889.......... 1889-1890.......... 1890-1891.......... 1891-1892.......... 1892-1893.......... 1893-1894......... 1894-1895.......... 1895-1896......... STRALSUND: 1888............... 1889.............. 1890............... 1891............... 1892............... 1893............... 1895-1896.......... 127,000 132.000 138,000 141,000 145,000 149,000 154,000 156,000 160,000 164,400 170,300 173,500 177.200 181,400 185,600 218,000 362,600 378,000 381,300 396,400 399,900 405,000 414,000 422,000 36,000 37,000 38,000 40.000 43,000 45,000 40,000 50,500 12,000 12,000 12,000 12,000 12,000 12,800 12,800 12,800 12,800 12,800 13,000 13,000 106,000 108,000 116,000 118,000 120,000 126,000 130,000 140,000 17,910 22.226 22,914 22,914 27,820 28.000 30,105 17.4 18.7 18.7 20.5 19.3 19.5 19.3 19.0 19.3 20.0 20.3 21.1 21.4 24.8 27.7 25.6 15.3 15.8 16.1 15.6 16.1 15.8 16.6 17.7 11.3 11.1 10.8 11.6 26.1 25.3 25.6 17.1 2,913 3,016 3,138 3,251 3,391 3,529 3,679 3,815 3,913 3,971 4,061 4,600 5,159 5,395 5,602 6,691 6,884 9,010 9,700 10,115 10,525 11,008 11,655 12,342 450 485 525 565 598 630 651 737 804 843 896 995 1,095 1,191 1,295 2,221 6,186 8,694 9,598 10,178 10,754 11,292 11,891 12,856 76 77 78 77 35 35 34 53 32 39 31 29 27 27 26 29 46 49 100 100 11 10 11 11 10 10 100 3 4 4 5 6 6 5 5 5 6 2.2 2.4 2.9 3.8 5.7 17.7 18.5 9.5 23 (e) 17(e 17 (e: 18 (e 17 (e: 17 (e; 22 (e) 18 (e: 7 9 4 4 3 3 3 3 3 1 2 2 7 7 6 7 7 8 6 6 2.8 2.1 2.1 2.1 1.9 1.9 7.6 1 1 2 1 1 1 Unfiltered water pumped. The per capita consumption was reduced one-half with the placing of meters to measure three-fourths of the supply taken by consumers, although in 1897 less than half the total supply was metered. Inhabitants per tap in 1897=30. Gravity supply. The per capita consumption iwas reduced one-half with the placing of meters ito measure two-thirds of the water taken by iconsumers, and still 10 per cent. more when all water was sold by meter, the percentage of the "total supply taken by consumers remaining practically the same during the twelve years. IThe amount of water used for public purposes ranged from 7.8 to 3.3 gallons per capita per day Filtered water since 1895. Pumped. This is one of the most important commercial cities in Europe. In 1899 about half the total supply was metered; of the metered portion about half was for domestic and the balance for manufacturing, trades, commercial and shipping consumers. The amounts used for public purposes and for the operation of the plant were comiparatively small; about 2 to 3 gallons per capita per day. The wastage is classed with the onmetered consumers' water. The result, since i11891, from which date the introduction of meters Ihas been at a more rapid rate than formerly, has Ibeen the reduction of this wastage.to the extent of about 11 gallons per capita per day, or 20 per cent., with about 50 per cent. of the total, and 21 per cent, of the domestic consumption metered. Inhabitants per tap in 1899=34. (b) Domestic consumers only. Unfiltered water pumped. The rapid intro6r> duction of meters on consumers during 1892 and 0.3 1893 was followed by the reduction of the con2:4 Isumption from 34 to 21 gallons per capita per 0.8 day; about 38 per cent., with 76 per cent. of the 8.8 total supply metered. The amount of water 11.9 used for public purposes has varied from 0.5 16.2 gallons to 4 gallons per capita per day. Inhabitants per tap in 1896=27. Unfiltered water pumped. Up to 1889 about.... 100 meters were added annually until approxi-.." mately one-third the services were metered. During this time the rate of consumption had '' been climbing steadily till it reached about 28.... gallons per capita daily in 1899. In 1890, 924.... meters were added, and by 1893 there were.... practically as many meters in use as there were.... services, and. the per capita consumption had... dropped to 16 gallons per capita, a reduction of.... 43 per cent. Number of inhabitants per service.... in 1898=34.:::: I:::: j It will be observed, in this case, that with the 1 (f) decrease in percentage of consumption metered 6 (f) in 1896-97 there was an immediate increase in 5 (f) the per capita consumption, followed by a sub5 (f) sequent decrease coincident with the increase in 48 (f) the percentage of metered consumption in 1899 -48 (f) 1900. This is one of the best illustrations that 44 (f) can be found of where waste is located. For the 29 (f) first four years in the table all services, except those for public purposes and for water consumed at the works, were metered and the amount of water unaccounted for is practically nil. Taking the meters off of half the consumers was followed by wastage equal at least to the former per capita consumption; and putting back half the meters that had been removed stopped about half the waste. (e) Unmetered. (f) Plant and waste, loss and unmetered consumers' water. 9.7 9.7 13.0 16.1 19.8 19.8 19.5 21.9 16.1 16.1 14.3 15.1 29 32 30 31 33 31 24 20 43.0 49.4 47.0 49.1 41.2 40.2 21.4 772 728 805 824 860 877 898 907 917 933 949 995 1,270 1,273 1,273 1,282 1,352 1,656 122 i30 156 159 226 228 231 413 916 937 976.:.. 223 233 233 233 236 243 1,656 11 11 12 12 13 21 45 76 11 10 11 11 10 10 60 91 89 94 95 95 95 95 95 95 96 69 69 92 92 93 92 92 91 91 93 96.7 97.9 97.9 97.9 98.1 98.1 60.3 2 2 2 1 2 2 2 2 2 3 29 29 1 (g 1 (g 1 (g 1 (g 1 (g 3 (g 1 (g... Ig (IS... (i... (I.. (g * *