mum rate of flow one-third greater, which would be represented by 200; and since the sewers must be proportioned to discharge the maximum flow occurring at any hour during the week, in this case it should be made to carry twice the average daily flow.

Changes of Temperature.-There are also wide variations in water consumption and consequently in sewage flow, due to climatic differences and variations in temperature, which are usually too little considered in proportioning the size of sewers, and particularly of main sewers.

In a majority of our cities, the maximum consumption of water occurs during the winter months, and as it is due to water taps which are left open to prevent freezing, a greater percentage of the total consumption than in ordinary cases reaches the sewers, and the maximum rate of water consumption becomes in still greater degree the maximum rate of sewage flow. The secondary maximum of water consumption, occurring in summer, is in greater degree used for purposes which withdraw it from the sewers, as in street and lawn sprinkling, etc.

It is during this maximum rate of flow, occurring in the winter months, that sewers are more likely to be subcharged than at any other time, and it should be carefully considered in proportioning the sizes of pipes in the system. It is true that it is the result of a waste of water, which is perhaps extravagant, and which it might be better economy to checkthan to remove by sewerage.

Use of Water Increasing. It is also true that the per capita consumption and waste of water has been gradually increasing up to the present time, and is likely to reach still higher figures. This increased demand for water has been met by pumping engines of much higher duty, and by improvements in water works generally, which enable them to furnish water to the consumer at lower and lower rates

per gallon, commensurate with the increased economy secured. This in turn encourages the use of water from the public mains for motive power, as the running of elevators, motors, etc., and for the thousand and one purposes of light manufacturing, requiring the use of power, always ready, and costing nothing when not wanted. Rapid as has been the development of water supply systems in the United States, their capacity has barely kept up with the demands of the people.

It is likely, however, that recent developments in the electrical transmission of power and a growing tendency to conserve the water supply by more rigid inspection and the general use of water meters, will tend to reduce the average consumption of water in most cases.

CHAPTER V.

QUANTITY OF SEWAGE.

Owing to the scarcity and incompleteness of data at present accessible on the actual flow of sewers, and to their unreliability as well, and to the very complete records of water consumption, which are made possible by the use of pumping machinery, automatically recording its own performances, an investigation and classification of the statistics of water consumption will, undoubtedly, be of use in designing a sewerage system. Size is entirely a matter of calculation from data, mainly assumed, as:

(1) The extent of the system.

(2) The density of population, or the probable density of population in the near future.

(3) The number of gallons of sewerage per diem per capita.

(4) The varying rates of sewage discharge.

(5) The inclination of the sewers.

(6) The smoothness of the interior surfaces of the

sewers.

Of the above factors entering into calculations determining size, those of extent and inclination only are capable of exact determination. And since, in designing a system of sewerage for American towns, the element of future growth must be considered, not only by increased density of population, but by the extension of the suburbs, thus extending the dead ends, and practically converting what was formerly a lateral into a main, the factor of extent may be considered to have an element of uncertainty.

Upon the accuracy of the assumption made by the engineer, then, in regard to these variable factors entering so largely into his calculations, will the efficiency of the system mainly depend.

The following statistics of water consumption are given with a view of showing what may be a proper value, for each of these variable factors in ordinary cases. They have been collected from various sources, and a summary is presented in condensed tabular form, convenient for reference. A few statistics of sewage discharge are also presented, but owing to the difficulties previously stated, they indicate the condition of sewage flow for a very brief period only, and admit of but limited application. When examined in connection with the statistics of water consumption they are of increased interest.

The consumption of water will be examined in the following manner:

(1) The quantity of water consumed.

(2) Monthly variations.

(3) Daily variations.

(4) Hourly variations.

(5) Variations due to extremes of temperature.

(6) Special cases-as, cities using water largely in manufacturing, brewing, etc.

In the following tables the standard of comparison taken is the average per diem per capita consumption, which is, for purposes of comparison, assumed at 100, and from statistical data the per cent. comparison is made by computation, as this is most convenient for use.

The average per diem per capita consumption is that most readily obtained, especially from pumping records of the smaller cities in which records are usually less perfect, being computed from automatic counter readings. In many records of pumping works, where the supply is com

pared with the population, the results are misleading. For instance: in the smaller cities of from 10,000 to 15,000 inhabitants, while the per diem per capita consumption, based on the total population, is but little below the average, the actual per diem per capita consumption for each person using water from the city mains must be greatly above the average, as the proportionate number of consumers in such cities is frequently below one-half, and, consequently, the sewage flow in proportion to the actual tributary population is disproportionately large. This condition is, undoubtedly, owing to the usual laxity of cities of this class in controlling the use of water, and is corrected as the city increases in population and improves in its conduct of municipal affairs.

The water statistics of large cities are often misleading in the other direction, many supplying water from the city mains to their own entire population, and to outlying suburban districts as well, which, in many cases, is not stated in published reports. In many cases, also, the per capita consumption in published returns is based upon estimates of population, which are merely guesses and may be wide of the mark.

The sewerage of every city presents problems for solution essentially peculiar to itself, and these must be carefully considered. The tables here given, while not strictly applicable to certain special cases, will, nevertheless, be a guide in determining their requirements.

The Quantity of Water Required.

The following is J.

T. Fanning's estimate for American cities: *

"In American cities having well arranged and maintained systems of water supply, and furnishing good, wholesome water for domestic use, and clean, soft water adapted to the use of the arts and for mechanical purposes, the average consumption is found to be approximately as follows:

(a) For ordinary domestic use, not including hose use: 20 gallons per capita per day.

*From J. T. Fanning's Hydraulic Engineering, by permission.