Springs.

Springs should be the most natural method of securing water-supply for a detached house, since no expense is involved except that of piping the water to the building. In Europe, spring water-supplies have been greatly developed in furnishing water for large cities. Vienna, for example, with its population of nearly two millions, obtains its water-supply from springs in the Alps mountains, and many smaller cities do likewise.

But in this country springs have been little used for water-supplies, partly because of the uncertain quantity furnished and partly because of difficulty in acquiring title to the water rights. If an individual, however, has on his farm, or within reach, a spring furnishing a continuous supply of water, it would seem quite absurd not to make use of such a Heaven-sent blessing. Care must be taken always that a spring is not contaminated by surface drainage, and for this reason, as with shallow wells, the wall surrounding the inclosed spring should be extended above the ground and made impervious to water for at least six feet below the surface. In some cases it may be wise to convert an open spring into an underground one, putting a roof over all and then covering with earth and sod. Figure 33 shows a type suggested by the French engineer, M. Imbeaux.



Very often a larger supply from a spring may be obtained by collecting into one basin a number of separate and smaller springs. A swampy or boggy piece of ground is often the result of the existence of a number of springs, and if drains are laid to some convenient corner of the field, and a well dug there, into which the drains will discharge, not only will the swamp be drained, but an ample supply of water in this way be obtained. It would, of course, not be wise to have cows pasture in this part of the field, nor, even when the ground has been dried out, should this field be manured or cultivated. It should rather be fenced and left to grow up in underbrush, dedicated to the farm water-supply.

Extensions of springs.



Again, if the water comes from a stratum W-W, as shown in Fig. 34, a large additional yield can be obtained by extending the spring from the point where it breaks out along the edge of the water-bearing stratum on each side. This extension or gathering conduit can be made by building rough stone walls on each side of the ditch, covering with flat stones so as to form a pervious channel to intercept the water and lead it to the chamber from which the supply pipe to the house leads out. The ground-water level will then be altered as shown by the broken line in the draining.

More simply it may be made by digging a trench along the hillside at the same level as the spring, or into the spring if necessary to find the water, and then laying draintile surrounded by coarse gravel or broken stone in the trench.

In the western part of the country much knowledge has been gained by investigating and experimenting on this kind of spring water development, only there the springs have been made artificially by digging down to meet the underground flow of water. For example, in the Arkansas River Valley, California, where it was suspected that water was flowing underground, a trench was dug transversely across the valley, and at a depth of six feet sufficient water was found to amount to 200,000 gallons per day for each one hundred feet of trench. On the South Platte River, near Denver, much the same thing has been done, and in a trench eighteen feet deep, water is collected at the rate of a million and a quarter gallons per day for each one hundred feet of trench. Other examples of the same sort might be given.

For a single house, the spring need usually only be extended by means of a short trench, and three-inch terra-cotta tile should be laid in the trench and surrounded by gravel and then covered over. The spring receiving water from these tiles should be inclosed, as will be described in a later chapter.

Supply from brooks.

Whenever a spring is not available and at the same time a supply of running water by gravity is determined on for a house, recourse is generally had to brooks which may find their way down the hillsides in the vicinity. In many instances the water in such brooks is practically spring water and is the overflow of actual springs. Where the brook is not subject to contamination between the spring and the point at which the supply is taken, the latter is as truly spring water as the former, and if a long length of pipe is saved, there can be no objection to the brook supply. On the other hand, it is suggestive, at least, of misrepresentation for a summer hotel or boarding house to advertise that their water-supply comes from springs when really it comes from an open brook miles away from the spring which may be indeed the origin of the brook, but with so many intervening opportunities for contamination that the pure original source is unrecognizable.

There are two obvious drawbacks to the use of brooks: (1) that the quality of the water is, in many cases, objectionable, and (2) that brooks are very apt to dry up in summer on account of their limited watersheds. The discussion on the first point will be postponed to a later chapter, and we have now to consider the question of quantity only.

The wisest plan before deciding on a brook supply is to measure the volume of water which flows in the brook at the time when it is lowest, probably about the middle of August. The actual volume of water needed for the household is not large, although its required rate of flow may be high and, as already pointed out, a stream which furnishes water at the rate of one quart in five minutes is sufficient for a family of three persons, a rate which is almost a drop-by-drop supply. Such a stream would require a reservoir somewhere in order to supply the faucets at the proper rate, and for a single family a small cistern or even a barrel sunk in the ground would be sufficient for this purpose. An objection to the utilization of so small a flow in connection with the smaller storage is that the temperature of the water in summer is so raised that vegetation and animal growths take place easily and freely, so that the taste and smell of such water is most disagreeable. These consequences can be avoided even with the low flow by increasing the storage, since the larger quantity of water has been found to resist the bad effects of the low flow and high temperature. Figure 35 shows a small reservoir actually in use to supply water for a single house.



Storage reservoirs.

But even if the stream actually dries up for two or three months, it is still possible to use it for water-supply, provided a suitable location for a dam and pond can be found where storage, as described in the preceding chapter, can be secured. For this reason as well as for the greater benefit to the quality of the water, brooks flowing through rough, wooded, and uninhabited country are to be preferred as a source of water-supply to brooks flowing through flat agricultural land, and in many cases, where their flow is largely due to springs, the brooks themselves may compare favorably with springs in quality.

Ponds or lakes.

Water may be properly taken from ponds or lakes whenever the danger from pollution is negligible. No better source of supply can be imagined than a pond in the midst of woods, far away from human habitation, presumably furnishing an unlimited supply of pure soft water. Sometimes water from such ponds contains large amounts of vegetable matter, the result of decomposition of swampy or peaty material, as, for instance, from the ponds in the Dismal Swamp of Virginia, so that the water has a yellow, coffee-colored appearance. The appearance of such water is suspicious, but it need not be feared unless something more pernicious than the coloring matter is present.

As the country becomes more settled, ponds are more and more likely to become contaminated and hence unfit for a water-supply, and this possibility must be taken into account in planning for a water-supply. It would be most shortsighted to carry a long line of pipe from a house to a pond several miles away, only to have the pond made unfit for use within a few years by the growth of the community around the pond. The possibility of cooeperation ought not to be overlooked, however. It is quite possible that half a dozen householders might be so located with respect to each other and to a pond that an arrangement could be made whereby the owner of a small pond would agree to fence it around and dedicate it to the purposes of a water-supply, doing this as his share. The others might then well afford to pipe the water to one house after another, including that of the owner of the pond.

Water from a pond or lake has one great advantage over water from a brook, namely, that contaminating substances in the pond settle out, so that pond water, especially if the pond is deep, is always of much better quality than running water. For this same reason, water taken from a reservoir on a stream is much better water than that in the stream above the reservoir indicates, and pollution is much less to be feared where the reservoir exists.

Pressure for water-supplies.

The value of a high pressure in the water-pipes of a house has been much overestimated. For a number of years the water-supply in the writer's residence came from a tank in the attic, the pressure in the bath-room being not more than ten feet, and while the water flowing through a three fourths inch pipe was noticeably slow, it was not so slow as to discredit the supply.

A height or head of twenty feet above the highest fixture in the house would be better and ought to be secured whenever possible. This head is obtained by having the source of supply higher than the highest fixture, not merely the twenty feet mentioned, but also an additional height necessary to offset the frictional losses caused by the running water. The loss from this source in case of fire supply has already been referred to, but for purely domestic supplies the loss is appreciable. The maximum rate as already indicated is not more than 7000 gallons per day, whereas the fire rate both for single houses and for a small hamlet is about a million gallons a day. For the lower rate, as well as for rates one half and twice this rate, the friction loss in vertical feet per 100 feet run in small pipes is shown in the following table:—

TABLE X. SHOWING LOSS OF HEAD BY FRICTION, FOR DIFFERENT QUANTITIES OF FLOW, AND IN DIFFERENT SIZES OF PIPES

======================================================================== Rate of Flow in Gallons Per Day 1/2" Pipe 5/8" Pipe 3/4" Pipe 1" Pipe 1-1/4" Pipe -+ -+ -+ -+ -+ - 3500 13.95 4.81 2.35 0.66 0.25 7000 47.17 17.30 7.45 2.04 0.74 14000 163.09 57.8 25.00 6.64 2.41 ========================================================================

The table shows how much additional elevation is needed over the 20 feet already referred to. For example, suppose it is decided that a rate of 1 quart in 10 seconds is to be maintained from three faucets or a rate of 7000 gallons per day. Suppose that a pond 4000 feet away is found to be 50 feet above the highest faucet in the house, and it is a question what size pipe ought to be used. By the table a 1-inch pipe loses 2.6 feet per 100 feet or 104 feet in the 4000 feet, an impossible amount when only 50 feet are available, although the size would be entirely proper if the difference of level was 124 feet or anything greater. A 1-1/4-inch pipe, however, loses only 0.74 foot in 100 or 39 feet per mile, so that the 1-1/4-inch pipe would be necessary, although that size would answer even if the pond were a mile and a quarter away.

When water from a well is pumped to an elevated tank there is the same necessity of providing about 20 feet difference in level between the tank and the highest fixture, but the length of pipe involved being small, the friction losses are not great. It should be noted even here that too small a pipe may reduce the pressure, a 1/2-inch pipe causing a loss of 47 feet in a 100-foot pipe line. If a tower is built by the side of the house, the distance down to the ground, across to the house, and up to the second floor would hardly be less than 50 feet, and this is a loss of 23-1/2 feet, which means that the tank would have to be set higher in the air by this amount. With a 3/4-inch pipe, it should go 3.7 feet, and with a 1-inch pipe but a foot higher than the level necessary to make the water flow out of the faucet at the rate already specified.



CHAPTER VII

QUALITY OF WATER

A pure water-supply has always been regarded as desirable and its value can hardly be overrated, from the standpoint of health, happiness, or economy. From the earliest history, no crime has been so despicable as that of deliberately poisoning a well from which the public supply was obtained, and in the past no charge more quickly could stir the populace to riot. In Strassburg in 1348 two thousand Jews were burned for this crime charged against them; and as late as 1832 the Parisian mob, frantic on account of the many deaths, insisted that the water-carriers who distributed water from the Seine, shockingly polluted with sewage as it was, had poisoned the water, and many of the carriers were murdered on this charge.

Yet no water, as used for drinking purposes, is absolutely pure, according to the standards of chemistry. Distilled water is the nearest approach to pure water obtainable, and it is said by physicians that such water is not desirable as a habitual and constant beverage. The human body requires certain mineral salts particularly for the bones and muscles, and while these salts are provided in a large measure by food, a number are also furnished by drinking water. On the other hand, a wonderful natural process is accomplished by distilled or approximately pure water in that the water tends to dissolve, to add to itself, and to carry away whatever excess of solids may exist in the body. For certain kidney diseases, for example, pure water is prescribed, not merely as a means of preventing further accretions, but for the purpose of dissolving and removing the undesirable accumulations already existing.

Practically, considerable latitude is possible in the matter of the purity of drinking water, and no particular harm is to be apprehended by the constant use of either a water containing as little as ten parts per million of total solids or of water containing as much as three hundred parts per million of total solids. The human body, in this as in so many other ways, is so constituted as to be able to adjust itself to varying conditions of food, and, until an excessive amount of ingredients are absorbed, no great harm is done. There are, however, certain definite substances—animal, vegetable, and mineral—which, when found in water, are decidedly objectionable, and it is not the amount of foreign matter in a water-supply, but its character, which is of importance in a water to be used for drinking.

Mineral matter in water.

The mineral matter is the least objectionable as it is also the most common, since all water is forced to partake, more or less, of the nature of the rocks and soil over which it passes. Good waters contain from twenty to one hundred grains per gallon of mineral salts; that is, of various chemical substances which are able to be dissolved by water. If the amount is much in excess of one hundred parts, the water is noticeably "hard," and this may increase to a point where the water cannot be used. For example, the writer once superintended the locating and drilling of a well which passed through a bed of sodium sulphate or gypsum, just before reaching the water, so that as the latter rose in the well it dissolved and carried with itself a large amount of this salt, so much that the water was useless. Water containing more than one hundred grains per gallon of such salts as magnesium sulphate or sodium phosphate is a mineral water rather than a good drinking water, and while an occasional glass may do no harm or may even have desirable medicinal effects, such a water is not fit for constant drinking.

It is worth noting that many attempts have been made to show the relative effect of various hard waters on the health. A French commissioner reported that apparently people in hard-water districts had a better physique than in soft-water districts. A Vienna commissioner also reported in favor of a moderately hard water for the same reason. It is to-day believed by many that children ought to have lime in water; that is, ought to drink hard water to prevent or ward off "rickets" or softening of the bones. An English commissioner, on the other hand, has concluded that, other things being equal, the rate of mortality is practically uninfluenced by the softness or hardness of the water-supply. This same commissioner has also shown that in the British Isles the tallest and most stalwart men were found in Cumberland and in the Scotch Highlands, where the water used is almost invariably very soft (Thresh's "Water-supplies").

It has been asserted that certain diseases, not necessarily causing death, are caused by hard water, as calculus, cancer, goiter, and cretinism; but, as already pointed out in Chapter II, no satisfactory proof has ever been established. One must conclude that within reasonable limits there is little to choose between a hard and soft water for drinking purposes, although a change from a soft water to a hard, or vice versa, usually produces temporary derangements.

Loss of soap.

For washing purposes the value of a soft water is more marked. When a hard water is used, a certain amount of soap is required to neutralize the hardness before the soap is effective, and this takes place at the rate of about 2 ounces of soap to 100 gallons of water for each part of calcium carbonate per gallon, or about 3 ounces of soap to 10,000 gallons for each part per million increase in hardness.

The village of Canisteo, New York, has a hard spring water, the hardness being recorded by the State Department of Health as 162.8 parts calcium carbonate in a million parts of water. Clifton Springs water has a hardness of 208. Catskill, New York, which gets its water from a stream running down from the hillside, has a hardness of 22.1 or 140.7 parts less than Canisteo. Mr. G. C. Whipple says ("Value of Pure Water") he has found that 1 pound of soap is needed to soften 167 gallons of water when that water has a hardness of 20 parts per million, and that each additional part requires 200 pounds of soap to soften a million gallons. If Clifton Springs and Catskill should each use 100,000 gallons per day, the additional cost of the hard water, at five cents a pound for soap, would be 20 x 140.7 x 0.05 = $140.70, provided all the village water were neutralized with soap. Probably not over one fiftieth part of the water is so neutralized, so that the added cost of soap is actually about $2.80 a day. Whipple expresses this cost as H/100 = D, where H is the hardness in parts per million and D is the cost in cents for every 1000 gallons used for all purposes. Thus Canisteo water costs 162.8/100 = 1.6 cents per 1000 gallons used, while Catskill costs only 22.1/100 or 0.2 cent on account of soap.

This discussion is intended to suggest a comparison between a well of hard water and a surface supply of soft water, when both are available. It should arouse an interest in securing a soft water as well as a clear water, and the advantages of the softer water, in so far as soap consumption alone is concerned, are seen to be not inconsiderable.

Vegetable pollution.

The vegetable and animal matter is organic in its origin and nature, and their effect on water may be taken up together.

Vegetable pollution is generally the result of decayed leaves, roots, bark, and such other vegetable tissue as would be likely to be found where the water-supply flows through a swamp or accumulates in hollows and depressions. This sort of water is likely to have a brownish or yellowish brown color, to have a slightly sweetish taste, and to be soft, that is, free from mineral solids. Usually such water can be used for drinking purposes without serious consequences. AEsthetically, it is objectionable because of its color, and the city of Boston has expended many thousands dollars in building channels around swamps and in providing artificial outlets for swamps, so that the color of the water collected on the watershed shall not show the color induced thereby. Water from the Dismal Swamp of Virginia is so discolored as to look like coffee, and yet, in the vicinity, it is much prized for drinking, and formerly great pains were taken to fill casks with this water when in preparation for a long sea voyage.

Such matter always has a marked influence on a chemical analysis of the water, shows large amounts of nitrogenous matter, and apparently indicates a polluted supply; but, if the reason for this apparent pollution lies in the presence of a swamp, no danger to health therefrom is to be apprehended. Such water also is less subject to decay or putrefaction, and if a water-supply for a house is to be taken from a small pond, a gathering ground containing swamps is likely to furnish a more satisfactory water, color alone excepted, than one free from such swamps.

Pollution of water by animals.

Animal pollution usually comes from the presence on the watershed of domestic animals, that is, cows, sheep, and horses, or from manure spread on fields draining into the brook, or from barns or barnyards close by the water. It is the presence of this sort of pollution that furnishes the other kind of organic matter not to be distinguished by chemical analysis from the organic matter just referred to, but vastly more objectionable.

Drainage from houses and barns is responsible for the same kind of animal pollution, and while it is difficult to prove by statistics that such pollution is always dangerous to health, it is sufficiently repulsive from an aesthetic standpoint to be done away with whenever possible. Such pollution applies only to surface water, such as brooks or lakes, and the best method of detecting and evaluating this pollution is to make a careful inspection of the watershed.

If it is proposed to use the water from a certain stream for drinking purposes, the first step should be to examine carefully the area draining into the stream, to detect, if possible, all opportunities for animal wastes to find their way directly into the stream and to note whether fields sloping rapidly to the streams are manured; to see whether the stream flows through pasture land in which cows are kept, and especially to note whether houses with their accompanying outbuildings are near enough the brook so that water may at any time wash impurities down into the stream. Whenever a brook flows through woodland free from all animal pollution and not subject to pollution before entering the wood, the water is probably as pure as that in any spring or well.

On the contrary, when the water in a brook flows through a meadow used for pasture or through gullies, the sides of which are manured, or in the vicinity of houses and barns, the water is probably unfit for drinking purposes. This can be realized by standing at the edge of a barnyard and watching the rain falling first on the roof of the barn, then in larger quantities from the eaves on to the manure pile into the yard below, then accumulating in pools of reddish black concentrated liquid, until the volume is sufficient to form small rills which gradually assemble into a fair-sized stream. Similarly, the pig-pen drainage is washed out from under or even through the building, and, after combining with the barnyard drain, is carried into the stream near by. The very idea of drinking such filth is nauseating in the extreme. It is common for small slaughter-houses to be built on the side of a stream, so that the offal, carrion, and refuse of the place may be carried off without effort on the part of the owner, and there are a number of such places where brooks, used as places of deposit for slaughter-house refuse, discharge directly into the reservoirs of water works.

But this sort of animal refuse is not the most serious pollution. The leachings and washings from privies and cesspools, carrying, as they do, germs of contagious diseases, are most to be dreaded, and when a privy (with no vault underneath) is built on the side of a steep ravine and is so located that the natural drainage of the sidehill on which it is built cannot help but run around and through the building, then the pollution of the stream in the gulley is not only direct and inevitable, but of a deadly sort (see Fig. 36). Fortunately, the germs thus carried into the stream suffer the vicissitudes of all life exposed to the attacks of hostile forces.

At the time of freshets the streams carry mud in abundance, which mud is continually settling out of the water as opportunity offers, and with this settlement of mud there occurs also the settlement of the germs. Also the pathogenic or disease-producing germs are usually weaker and more susceptible than the putrefactive and other organisms which are found in the water in great abundance after any rain storm, and which tend to inhibit or destroy the pathogenic germs. But some will survive, and, with favoring conditions, may pass through the water-pipe to the house, causing sickness, if not death.



Any inspection of the watershed, therefore, should look to the elimination of the dangers above described, and to the location of barns and barnyards, pig-pens and poultry yards, privies and cesspools, so that no direct drainage into the stream shall be possible.

It is out of the question for any surface water-supply to be pure, since the mere fact of the passage of water over the soil inevitably results in the collection of organic matter; and it is no exaggeration to say that the time will inevitably come in this country, as it has already in Germany, when no surface supply will be considered satisfactory unless the water is filtered. The only alternative is water gathered from areas that are owned by the individual and on which, therefore, all dwellings may be prohibited, all cultivated land avoided, and where the primeval forest may be restored, making the watershed equal to that from which forest streams emerge.

But usually, in the case of a single house, it will not be possible entirely to eliminate the dangers of surface pollution, although an inspection will show the dangers, and possibly some of them may be avoided. Certainly any direct drainage into the streams should be cut out, as well as the drainage from barnyards in the immediate vicinity of the point where the water is taken out. Just what percentage of pollution may be eliminated in this way it is impossible to determine, but it is not too much to say that no brook or pond should be used for a water-supply of a house unless every known pollution of an organic nature has been removed. Under the most favorable circumstances there will be enough accidental contamination to make the water at times dangerous, and no added risks ought to be assumed.

In looking over a watershed the possibility of sewage entering the stream is, of all pollutions, the most to be avoided. To adequately investigate the quality of a stream, the inspector must satisfy himself as to the point of discharge of the sewer of every house on the watershed, and this must be done personally, without apparently reflecting on the statements of the owner of the house. If any such points of discharge are found, the sewage should be either diverted into some other watershed, or spread out over the ground away from the stream, or purified by some artificial treatment before discharge, or else the creek water cannot be used.

The next point to be noted in the source of the water-supply is the presence and location of privies. These nuisances should be as far back from the banks of the streams as possible to eliminate all danger since the surface of the ground always slopes toward some stream, and pollution may be carried for considerable distances over or through the soil. Water-tight boxes can be provided so that no possible pollution of the surface-wash can occur, and if periodically the contents of these boxes be hauled away and buried, the privy loses its dangerous character. The city of Syracuse has installed on the watershed of Skaneateles Lake a most admirable system of collection of privy wastes, and the lake water is thoroughly protected, although there are several hundred privies on the watershed.

Cesspools, in general, are not dangerous if they are located fifty feet or more from the stream and if no overflow occurs.

Barnyards ought not to drain directly into streams, but when, as in so many cases, the stream flows through the barnyard, the only remedy is to move either the stream or the barnyard, and it is difficult to persuade even a well-disposed neighbor to do either. It is sometimes possible to appeal to his sense of right; but, too often, the neighbor feels that it is his land, his barn, his drain, even his brook, and he will do whatever he pleases with them, whether the water further down stream is to be used for drinking purposes or not. The question resolves itself into an inspection of the watershed and a determination of the existing conditions. If those are tolerable, the water may be used. If evident contamination is present, the water must usually be given up, and some other source of supply sought.

Well water.

The pollution of wells, if it exists at all, is usually very pronounced, and it is probably safe to say that, except where buildings, drains, or cesspools have been crowded too close to wells, or where some manifest and gross cause of pollution exists, a well water is safe to drink.

To protect properly a well from gross pollution, two precautions should be observed. The wall of the well should be built up in water-tight masonry, so that surface wash cannot enter the well except at a depth of at least six feet, and second, this water-tight masonry should be carried above the surface of the ground at least six inches and the well then covered with a water-tight floor so that no foreign matter can drop through the floor into the well or can be washed in by the waste water from the pump (see Figs. 28, 29, 30). If these precautions are taken, it is safe to say that nine tenths of the pollution occurring in isolated wells would be stopped.

Besides the above, a well may be polluted by a stream of underground water washing the contaminating matter through the soil. Experiments have been made to show this very plainly. A large number of bacteria were placed six feet below the surface just in the top of the underground stream of water. Within a week they were found in considerable numbers in the water of the soil one hundred feet distant, but when the same number of bacteria were placed in the soil four feet below the surface above the level of the ground water, none of them found their way into the water of the soil. This experiment shows the folly of building a cesspool in the vicinity of a well when they both go down to the same water level, since the contents of the cesspool will be carried into the well if the underground stream flows in the proper direction. A shallow cesspool, however, would not be open to the same objection.

It is always difficult to detect the direction or flow of underground water, and various technical and delicate methods have been selected to make this determination. A very simple test, however, is to dig a hole at the point where pollution is suspected, carrying the hole down to where ground water is reached, and then to throw a gallon of kerosene oil into the hole, and if the ground-water flow is toward the well, the presence of kerosene in the well water will make the fact known. This would not, however, prove that the actual contamination would produce disease, since a liquid like kerosene can find its way through the pores of the soil to much greater distances than bacteria can be carried. But, to be on the safe side, water from such a well should not be used.

To make sure of the quality of the water proposed for a water-supply, it is wise to have such water examined by a chemist. The chemist will make certain determinations of ammonia and other chemical combinations, and will report his findings with an interpretation or explanation of the result. What he finds is not the presence or absence of disease or disease germs, but substances that suggest or involve the presence of organic pollution. A test is made for the number of bacteria, and a well of spring water which contains more than about fifty in a cubic centimeter is a suspicious water. Surface water, on the other hand, may contain two or three hundred without being necessarily bad, the types of bacteria being harmless. Generally, a chemist will also determine the presence of the colon bacillus which is found in the intestinal tract of man or warm-blooded animals. Wherever this is found, in even such a small quantity as one cubic centimeter of water or less, there is strong presumption that the water has been polluted by human wastes and is therefore not fit to drink.

Dangers of polluted water.

Since no evidence of the danger of drinking polluted water can be so graphically expressed as by a direct reference to epidemics caused by the unwise use of such water, it will not be out of place to refer briefly to some of the instances in which a direct connection has been traced between a specific pollution of a certain water and disease or death resulting from it.

Although, as has already been explained, an infected water causes various kinds of intestinal disorders, particularly among children, the most characteristic evidence of pollution occurs when the noxious material comes directly from a typhoid fever patient, so that this same disease can be recognized as transmitted to another individual or family. This transmission of typhoid fever, while in some cases very plainly due to other agencies than water, as, for example, milk, oysters, and flies, yet, by far the largest proportion of the transmitted cases comes through the agency of polluted drinking water, and there are many examples both of contaminated wells and streams which emphasize this possibility beyond all question.

Two historic investigations of epidemics which have thoroughly convinced sanitarians that typhoid fever is a communicable disease and that water is the vehicle for its transmission may be briefly cited.

In 1879 Dr. Thorne reported an epidemic in the town of Caterham, England, which he had investigated, and disclosed the following facts: The population of the village was 5800. The first case of fever appeared on January 19. Others followed in rapid succession, until the number reached 352, of whom in due time 21 died.

The possibility of infection was carefully looked into. The influence of sewer air was ruled out because there were no sewers. The milk supply was proved unobjectionable. No theory of personal or secondary infection could account for the widespread prevalence, particularly as only one isolated case had occurred during the preceding year, and this had been imported.

Of the first 47 persons attacked, 45 lived in houses supplied with the public water-supply, and the other two were during the day in houses supplied with public water. Further, in the Caterham Asylum, with nearly 2000 patients, not a single case appeared, their water coming from driven wells. Investigation of the water-supply showed the undoubted cause of the epidemic. The public water-supply was derived from three deep wells, connected by tunnels in the chalk. In one of these tunnels, from January 5 to the end of the month, a laborer worked, who, though unattended by a physician, was evidently suffering from mild typhoid fever, the symptoms of the disease being carefully detailed by Dr. Thorne. The laborer at the time of his going to work had a severe diarrhoea, and while in the tunnel was obliged to make use of the bucket, in which the excavated chalk was hauled to the top. He admitted that at times the bucket, in being hauled up, would oscillate in such a way as to spill part of its contents and thereby pollute the water of the well below. Two weeks from this accidental pollution the epidemic began, and there can be little doubt of the relation of this mild case of typhoid to the epidemic which followed.

A second illustration may be cited at Butler, Pennsylvania, which occurred in 1903. The water-supply of Butler, a borough of 16,000 people, comes from a reservoir on the creek which flows through the phase. On account of the gross pollution of the water at the pumping-station, a long supply pipe has been laid from the reservoir directly to the pumps. The water also was filtered through a filter of the mechanical type. Through some accident the filter was thrown out of service for eleven days, between October 20 and 31, 1903, and unfortunately, on account of the failure of the reservoir dam, the water was at that time being taken directly from the creek at the pump well, and had been since August 27. Only ten days after the filter was shut down, the epidemic broke out in all parts of the town. Between November 10 and December 19 there were 1270 cases and 56 deaths. In the subsequent investigation it developed that not only was the stream generally polluted by the sewage at various points above the intake, but that there had been several cases of typhoid fever on the watershed, some on a brook that enters the creek within one hundred feet of the filter plant. As at Caterham, the inference is patent that the introduction of some specific infection into the drinking water was the direct cause of the general epidemic.

The occasional outbreaks of typhoid fever which occur in single families are not so easy to explain, particularly since the small number of persons affected does not usually call for a widespread interest on the part of those experienced in such epidemics. In the Twenty-seventh Annual Report of the New York State Department of Health, the following description of an outbreak in a small hamlet, where the cause seems to have been the use of a pond for a wash tub by some Italian laborers, thereby transmitting the disease germs from their clothes to the water afterwards used in a creamery, is given. The diagram, Fig. 37, shows that the creamery secured its water for the purpose of washing cans from a small pond by means of a gravity pipe line. The foreman of the creamery, who boarded at the residence marked A, first contracted typhoid fever. A week later an employee at the creamery also contracted the fever, the residence of the latter being marked B on the diagram. About six weeks later the railroad station agent, living at the point marked C, contracted the fever, and two weeks later his wife was attacked with the same disease. The residences at B and C are only about three hundred feet apart, both families taking their water-supplies from a spring between the two, but nearer B. During the summer previous to this outbreak a gang of Italian laborers, engaged in double-tracking the Central New England Railroad, were housed in box cars standing on one track of the railroad. One of the members of the gang was reported to have been taken ill with a fever and was at once removed, it was supposed, to a hospital in New York. It was the practice of the Italian laborers to bathe and wash their clothes in the upper of the two ponds from which water is supplied to the creamery by the pipe line. All the persons who contracted the fever were supplied with milk from the creamery. The foreman, who was the first to contract the fever, used water from the creamery and from the well at the house where he boarded. The other families, as already mentioned, used water from the spring. The conclusions, therefore, are that the creamery in some way became infected with typhoid fever, probably through the water-supply from the pond, and that the first two cases were due directly to this cause; that the station agent and his wife contracted the fever because of the infection of the spring, either from some small stream which is the outlet of the ponds or from some infection due to the illness of the owner of the house B near by. The report concludes as follows: "The use of water for creamery purposes from a pond exposed to such unwarranted and unchecked pollution as is shown here, or the permitted abuse of a water-supply for a creamery, appears little less than criminal negligence on the part of those responsible for the management of the creamery."



Another report in this volume of the New York State Department of Health illustrates very well how a spring or well may be contaminated, and is taken from a report on an outbreak at Kerhonkson, Ulster County. The report reads as follows: "The village of Kerhonkson is built mainly on the side of a mountain of solid rock covered by a thin top soil of variable depth. Owing to its rocky nature, only one or two wells exist throughout the whole place; such a thing as a drilled well has never been seriously considered.

"The inhabitants obtain their drinking water from a well on the property adjacent to and above the present school building, and known as the "Brown" well, and from a clear spring at the bottom of the hill in the rear of the village store and known all over the region as the Loundsbury spring.

"The school building is an old-fashioned two-story ramshackle affair with overhanging eaves, especially designed to obstruct light and darken the upper schoolroom. The building is in the center of a pine grove 250 x 150 feet in size, which also obstructs the light and tends to dampen the building. At the extreme ends of this school lot are two privies for the boys and girls, built on loose stone foundations, innocent of mortar or cement, which allows the water in heavy storms to wash out the fecal contents of from nearly a hundred pupils down upon the habitations below. Were the wells existing in the village as carelessly constructed as the Brown well and the various privy vaults which I have inspected, the loss of life from typhoid fever would be terrible indeed.

"Obtaining the names of all the patients who had suffered from this disease, I found that all but three were Kerhonkson public school pupils, and all had drunk the water of the before-mentioned well on the Brown property. Two out of these three cases were mothers of pupils who had been stricken with the fever and who had nursed the children through their long and exhausting illnesses and afterward had been attacked by the disease themselves, while the third and remaining case was a puzzler. This boy had never been a pupil of the school in question, nor had he partaken of any of the water of the suspected well. He was a pupil of another school entirely and lived in an adjoining village a considerable distance away. A special visit to him, however, developed the fact that some time before his illness he had come to the village store in Kerhonkson to purchase goods and had drunk water from the Loundsbury spring.

"Two years ago two cases died of typhoid fever on the property on which the Brown well is situated. Their stools were treated with lime and buried on the hill behind the house. Three cases of the same fever have occurred in the same house this season. The well in question is laid up with stone and cement and was supposed to be tight and impervious to surface water contamination. Investigation, however, proved that there were openings in the stone work in the side toward the privy. On examining the privy it was found that the foundation was composed of loose stones without cement or mortar that would readily allow the fecal contents to be washed down toward the well, the privy being about three feet higher than the well, the natural descent of the land being about one foot in twenty-five, the distance between privy and well being only about eighty feet. Another factor favoring the well contamination from this privy is that any filth washed downward from the privy toward the well would be stopped by the wall of the house proper and carried directly toward the well which lies close to the southeast corner of the house. Thus all of the conditions point to privy contamination of this well which should be at once cemented up on the inside, thoroughly cleansed and purified, before its use should be permitted, while all the privies in question should be provided with vaults of brick eight inches thick with eight-inch brick floors all laid with cement, and their inside surfaces lined with cement at least one inch thick, to prevent any further possible contamination."

In view of the imminent danger always possible wherever human wastes are directly discharged into streams, whether from privies or sewers, it is obvious that water so contaminated should never on any account be used as drinking water. It does not follow, because a stream so contaminated has been used for months or years without producing any evidence of disease, that the water is safe. Unless an excessive amount of organic matter is so transmitted, no evidence will be found that such pollution has existed through any outbreak of disease. But if once the discharges become affected through a person having typhoid fever, then the result of the infection is apparent immediately. If, therefore, an inspection of the stream above the point where it is proposed to take the water-supply shows the existence of privies, as shown by Fig. 36, the water should not be used for domestic supply, although a number of individuals may have been using the water for years without bad effects. It is a case in which prevention is much wiser than cure, and while economy and convenience may indicate such a polluted stream to be a desirable source of supply, a proper regard for health conditions will rule it out absolutely.



CHAPTER VIII

WATER-WORKS CONSTRUCTION

Construction methods and practices which lend themselves to the development of the water-supply for an individual house may be divided into three parts, namely:—

(1) Construction at the point of collection, whether this point be a well, spring, brook, or reservoir;

(2) The pipe line leading from the collection point to the buildings;

(3) Constructions involved in the house, other than the plumbing fixtures.

Taking up these different points in order, we may note at the outset that it is possible to employ either very simple or very complicated construction.

Methods of collection of water.

The common method is to lay a galvanized iron pipe in a ditch as far as a spring and there to protect the end of the pipe with a sieve or a grating and to leave it exposed in the water with no efforts expended on the spring itself. In a brook with waterfalls or with good slope, it is not uncommon to project a large pipe or a wooden trough into the stream at the top of a waterfall and so carry a certain amount of the water into a tub or basins from which the small pipe leads to the house. On the shores of a lake or pond the galvanized iron pipe is laid out on the bottom of the lake with the end protected by a strainer.

In all these cases the simplest method is the best, provided the supply of water is not needed in the winter; but such simple methods as just described fail when frost locks up the surface flow of the stream. Then the pipe throughout its entire length must be in a trench below the frost line at the entrance to the spring as elsewhere. To permit this, the spring must also be deep, or else so inclosed that the pipe leading into the spring can be covered by earth banked up against it. Not long ago the writer saw a pipe taking water from a small lake recently improved by a stone wall. Instead of conveying the water-pipe down under the wall the unwise stone mason had built the wall around the pipe and the pipe line was frozen up through the entire winter following.

Such simple methods also fail when the supply of water is not adequate, since, in order to secure a large quantity from a stream whose flow is periodic and irregular, some storage must be provided, and storage usually requires more or less elaborate construction work at the reservoir. Another reason for more elaborate construction at a spring is to prevent surface contamination, and it is always desirable to roof over a spring in order to protect it from surface flows. The writer has seen, as an example of objectionable construction, a spring in the bottom of a ravine or gully down which, in time of rain, torrents of water passed, although in a dry season the spring was the only sign of water in the vicinity. It could not but happen that this torrent of water, which carried all kinds of pollution from the road above, practically washed through the spring, destroying its good quality. In such a case, another channel for the gulley water ought to have been made, or else the spring dug out and roofed over, so that the torrential water could pass above it.

In other cases, the spring is found at the lowest point in a general depression, so that, while no stream passes through the spring, the spring is a catch-all for the surface drainage in the vicinity. In such cases the water should be protected by a bank of earth around the spring, behind which the drainage should be led off through a special pipe line if necessary.

Spring reservoirs.

In protecting the spring and in building up around it in order to put it underground, concrete is the most suitable material, although a large sewer pipe or a heavy cask or barrel will answer the purpose. It is usually sufficient to dig out the spring to a depth of four or five feet, and with a pump it is possible to keep the water down, so that the concrete walls may be laid. In building these walls, it is important to notice from which side the spring water comes, and on that side holes should be left in the wall. These openings may properly be connected with agricultural tile drains laid out from the spring in different directions, serving both to drain the ground and to add volume to the spring. It is often possible instead of pumping out water during construction to drain a spring temporarily, in places where the ground slopes rapidly, by carrying out a drainpipe from the lowest level; this drain is to be later stopped up.

The size of this spring reservoir depends on the average rate of flow of the spring and on the quantity of water used. If there is always an overflow from the spring, that is, if it always at all times of the year furnishes more water than is required by the house at that time of day when the greatest demand is made, then a two-foot sewer pipe is just as good as a concrete chamber ten feet square. But if at times the spring is low, so that the flow during the night must be saved to compensate for the excess consumption during the day, or if the rate at which the water is drawn at certain hours is greater than the average rate at which the spring flows, then storage must be allowed for in preparing the spring to act as a reservoir.

We have already estimated that a family of ten persons might use five hundred gallons of water a day, and the most exacting conditions would never require the spring to hold more than one day's supply. This would mean a chamber four feet deep and in area four by five feet. If the average supply of the spring is less than the average consumption of the family, then the spring must become a storage basin for the purpose of carrying water enough over the dry season, and the capacity of the basin must be computed from the number of days' storage required. It may not be out of place to suggest again the possibility of increasing the yield of the spring by laying draintile in a ditch running along the permeable stratum. These pipes may run fifty or one hundred feet each way from the main spring, so long as they continue to find ground water.

The walls of such a spring reservoir as here suggested for depths of six to eight feet need not be more than nine inches thick, whether built of brick or concrete. For greater depths the thickness should be increased to twelve inches.



The roof of the spring-chamber may be of plank, but this is temporary and undesirable. It is far better, for all spans up to ten feet, to make the roof a flat slab of concrete six inches thick, imbedding in the concrete in the bottom of the mass some one-half-inch iron rods, spaced about a foot apart each way and extending well into the side walls. The size of these rods should increase with the size of the chamber, making them three-quarter-inch rods up to a nine-foot span, and one-inch rods up to a twelve-foot span. There should be some way of getting into the spring, preferably by an opening in one corner so arranged as to carry the side walls of the opening or manhole up above the ground, where it may be protected with an iron cover locked fast (see Fig. 38, after Imbeaux). Besides the outlet pipe from the spring, which will naturally pass through the side walls about halfway between top and bottom in order to get the best water, there should be a drainpipe from the lowest part of the inclosure, the valve of which can be reached through a valve box coming to the surface. In the figure the drainpipe is shown by the dotted line, and the twofold chamber is for the purpose of allowing an examination of the spring to be made at any time.

The concrete used in this work should be of good quality, one part of cement to five parts of gravel or to four parts of stone and two parts of sand. A concrete bottom, although sometimes used, is not necessary. The position of the drain, of the house pipe, and of the several collection pipes must not be overlooked when the wall is being built, since it is much easier to leave a hole than to dig through the concrete afterwards.

Stream supplies.

If the volume of a stream is more than enough for the maximum consumption, nothing is needed but to carry the intake pipe from the shore out under water and protect the end with a strainer. In this case, however, the stream may freeze down to the level of the strainer and even around the strainer, so that the supply of water in winter would be cut off. To avoid this possibility the intake pipe ought to be in a pool of water so deep that it never freezes, and this means sometimes creating a pool for this very purpose. If storage is to be provided, a reservoir must be built, and this intake pipe would naturally be placed at least two feet below the surface of the water.

Dams.

If the stream is not deep, or if there is not a pool of satisfactory depth, or if the minimum flow of the stream is not adequate for the maximum needs of the consumers, a dam across the stream becomes a necessity. There are two or three types of dams suitable for a reservoir on a small stream, and they may be described briefly.

A dirt dam is not generally desirable, since in most cases the dam must also be used as a waste weir; that is, the freshets must run over the dam. This means that unless the crest of the dam is protected with timber or masonry the dam will be washed out; as happened, indeed, in the terrible flood at Johnstown, Pennsylvania, several years ago. If it is possible to carry the overflow water of the stream away in some other channel than over the dam, then a dirt dam is not objectionable, although always a dirt dam is best with a masonry core. A very good dam can be made by driving three-inch tongue-and-grooved planking tight together across a gulley and then filling in on each side so that the slope on each face is at least two feet horizontal for every foot in height. This last requirement means that if the dam is ten feet high, the width of the dam at the base shall be at least forty-five feet, the other five feet being required to give the proper thickness to the dam at the top.



In the second type of dam this central timber core is replaced with a thin wall of concrete as shown in Fig. 39, from six to twelve inches thick, sufficing to prevent small animals burrowing through the dam and at the same time to make the dam more nearly water-tight. Sometimes stone masonry is used, building a light wall to serve as the true dam, and then holding up this light wall with earth-filling on each side. If neither plank, stone, nor concrete can be used, the central core is made of the best earth available, a mixture of clay and sand preferably, and special pains are taken in the building to have this mixture well rammed and compacted.

The writer has recently heard of a dam on a small stream being made by the continual dumping of field stone from the farm into the brook at a certain definite place. This stone, of course, assumed a slope at each side and settled in place from year to year as the dam grew. The mud and silt of the stream filled up the holes between the stones, so that the dam was finally practically water-tight. This made a cheap construction and had the additional value of serving to use up stones from the fields. It was necessary, since the spring floods poured over the top of this dam, to protect the top stones, and a plank crest was put on, merely to keep the dam from being washed away.

The third type of dam is entirely of concrete or stone masonry, concrete to-day being preferable because more likely to be water-tight. The problem with a concrete dam is to get a foundation such that the impounded water will not leak out under the dam, imperiling the very existence of it. The ideal foundation, of course, is rock, and in a great many locations can be found in the small gulleys where the limestone and shale peculiar to this region will answer as well as more solid rock for dams not more than ten feet high; but with gravel banks on the sides or with soft sandy bottom, or where the clay soil becomes saturated with water at times, the gulley offers great difficulties for the construction of a dam. It will be wise, under such conditions, to carry a cut-off wall, not necessarily more than twelve inches thick, well into the bank, that is, about ten feet on each side, and under the dam this cut-off wall ought to go down until it reaches another stratum of sand or clay or rock. This cut-off wall, then, surrounding the main dam, shuts off the leakage, and the dam itself can be built without danger of undermining. In many large dams this cut-off wall is carried down more than a hundred feet, especially where the depth of water behind the dam is great. For small dams, a row of plank driven down behind a timber sill across and in the bed of the stream will often be sufficient.



The cross section of the main dam, in cases where flood water in the spring runs over the dam, should be such that the bottom thickness is about one half the height, and Fig. 40 (after Wegman) shows a suitable cross-section of a dam ten feet high. Figure 41 (after Wegman) shows a cross-section intended to carry the water over the dam, especially in times of flood, without danger of erosion.

Sometimes, in a narrow gorge with rock sides, it is possible to save masonry by building the dam in the form of an arch upstream, the resistance to the force of the water being then furnished by the abutment action of the rock sides, instead of by the weight of the dam, as in ordinary construction. For a dam ten feet high, the necessary thickness of the curved dam would probably not be more than twelve inches, while the ordinary gravity dam would be three or four feet thick. The workmanship on the former, however, must be of a very superior order.



It is never desirable to allow the water flowing over the dam to fall directly on the ground in front, since the falling water will rapidly carry away this soil and undermine the front of the dam. For this reason, the lower section of the dam is made curved, as shown in Fig. 41, giving the water a horizontal direction as it leaves the dam instead of a vertical. A plank floor is often added to carry even further from the dam any possible erosion (Fig. 40). Where it can be done, it is a good plan to provide a small body of still water below the dam, so that the force of the falling water may be distributed through the water on to the soil below.

There are other forms of dams often used. For example, brush dams, formerly common, are made by cutting off the tops of trees and dropping them in place and loading them with stones so as to make a mass of interwoven branches. These branches hold together particles of earth which are dumped in and form a dam.

Another dam that has been much used in rural communities is the old-fashioned crib dam, where logs are piled up crib fashion, held together at the corners by iron pins, a bottom spiked on, and the crib then filled with stone, a succession of these cribs across the stream forming the dam. Dirt is filled in on each side of this crib work, and, in some cases, cross timbers are set in, and both sides of the dam covered with tongue-and-grooved planking. But such dams are not permanent, and their construction involves an expense nearly equal to that of a permanent structure, and consequently they are not to be recommended.

Waste weirs.

When the dam is made of earth with or without a core wall and when no opportunity exists for carrying the waste water around the dam, a waste weir of masonry through the dam must be provided, so that freshets may be carried off without destroying or washing out the earth work.

The size of this weir is a matter of considerable concern, since its ability to carry off the high water is fundamental. The capacity of such waste weirs depends on the volume of flood-water, and this, in turn, depends on the area of the watershed. This volume cannot be predicted with any absolute certainty, but, in general, it may be said that the maximum run-off in the eastern part of the United States, from small areas not exceeding twenty-five square miles, will be about one hundred cubic feet per second per square mile, so that the freshet flow for a watershed of twelve square miles would be twelve hundred cubic feet per second. Ordinarily, the height of the weir is taken to be from two to four feet and the length made sufficient to care for the volume of discharge.

If the depth of water flowing over the weir is taken at one foot, the length of weir in feet necessary to carry the flood flow may be computed by multiplying the number of square miles of watershed by thirty. Then an area of twelve square miles would need a length of waste channel of three hundred sixty feet; in most cases, for small dams, longer than the dam itself.

If the depth be taken at two feet, then the number of square miles of watershed must be multiplied by ten to get the length of weir, so that a shed of twelve square miles would mean a weir one hundred twenty feet long.

The factor for a depth of three feet on the weir is six, making for the same area the length of weir seventy-two feet, and for four feet depth the factor is four. There is no more important part of the construction of a dam than that involved by a proper design of a waste weir, since a failure either to provide proper area or to so build as to withstand the erosive action of the running water will inevitably wash away the dam.

When the valley is narrow and the watershed large, the waste weir will occupy the entire width of the dam, and then it becomes necessary to construct the dam in masonry. On the other hand, when the watershed is small and the width of the valley great, then it is proper to make the waste weir only a certain portion of the entire width of the dam, making the rest of the dam either masonry or earth, as may be convenient.

Gate house.

In connection with a reservoir and at the back of the dam at the bottom of the bank, it is convenient to have what is called, in larger installations, a "gate house"; that is, a masonry or wooden manhole through which the water-pipe leading out from the reservoir passes and in which a gate is placed to shut off the water. In larger installations, it is usually possible to admit water at this point from different levels of the reservoir into the water-pipe, so as always to get the best quality of water, but for a small plant that is not necessary. A gate or valve, however, should always be provided, and while this may be on the bank of the pond with the intake pipe extending twenty or thirty feet into the pond, the valve should not be omitted. The end of the pipe extending into the pond should be placed about two feet above the bottom of the pond, instead of resting in the mud, in order to get a better quality of water.

Pipe lines.

In bringing the water from the spring or pond to the house, some kind of a pipe line must be provided. Such a pipe line is made of various materials; hollow wooden logs, vitrified tile, cast-iron pipe, wrought-iron pipe, and lead pipe having all been used. The last-named pipe is now too expensive for use in any great lengths. Hollow wooden pipes are employed occasionally, but, except in unusual localities, they also are more expensive than other forms, and are short lived on account of their tendency to decay. Cast-iron pipe, commonly used for municipal water-supplies, is not made in small sizes and may be excluded from the possibilities for an individual house. There remains only tile and wrought-iron pipe. Under certain conditions, the use of tile pipe is to be recommended, since it may be installed even in large sizes at a comparatively low cost, the objection to it being that it is very difficult to make the joints water-tight, and practically impossible when the pressure is greater than ten feet. It is more difficult to make joints in a pipe line of small diameter water-tight than in a pipe line of larger diameter, because the space for the cement in the former is so small. The writer has tried both four-inch and six-inch pipe, and while the four-inch line can be laid with tight joints, it requires much more careful and conscientious effort on the part of the workman than with six-inch pipe. The joints must be thoroughly filled with cement, not very wet, so that it can be rammed or packed with a thin stick into every part of the joint. Merely plastering the cement over the surface of the joint will always result in a leaking joint.

It often happens that a water-supply coming from a distance of a mile or so runs at first nearly level, so that, except for surface pollution, the water might be carried in an open ditch. An open ditch is, however, far better replaced by vitrified tile, six inches in diameter, which entirely prevents surface pollution, and which costs only about ten cents a running foot. When the slope of the ground exceeds the natural fall of the water, so that a pressure inside the pipe is created, iron pipe must be used. If vitrified pipe is used, the joints must be made with the greatest care, and every precaution taken to prevent leakage. Figure 42 shows a section of a joint in tile pipe.



In using iron pipe large enough to furnish the amount of water required, due regard must be paid to friction in the pipe. In flowing through a pipe of small size, water loses a great deal of head by friction. This friction between the sides of the pipe and the water, which must be duly considered in a pipe of small size, increases very rapidly as the velocity of the flow increases. It is always a great temptation to use a small pipe, since the cost of the pipe increases rapidly as the diameter increases, but it is penny wise and pound foolish to lay a line of pipe several thousand feet long to furnish water to a house and find when completed that the amount of water furnished by the pipe is on account of friction only a small dribble. In a previous chapter we estimated that the flow of water, in order to furnish three faucets at a reasonable rate, ought to be at least two thousand gallons a day or about one and a half gallons a minute, and the effect of a reduced size of pipe on the head necessary to carry a definite amount of water was shown.

The cost of cast-iron pipe should not be more than thirty cents per running foot for four-inch pipe and fifty cents per running foot for six-inch pipe. To this must be added the cost of about seven pounds or ten pounds respectively of lead for each joint and the cost of all the labor involved. The price of terra-cotta pipe is much less, as already indicated, so that it is quite worth while to expend some additional effort on making the tile pipe joints water-tight, if it allows the cheaper pipe to be substituted for the more expensive iron pipe.

Pumping.

Although the present methods of securing water for isolated farm buildings will not corroborate the statement it is safe to say that the proper method of obtaining a water-supply is always to make use of a pond or stream at such an elevation that water will flow to the house by gravity, provided this is possible. Only when the conditions are such that a gravity supply is impossible and water from a well or stream at some lower elevation becomes inevitable is pumping properly resorted to.

The advantage of a gravity supply is twofold. First, the daily charges for maintenance are practically nothing, so that when once the intake and the pipe line have been installed, there will be no additional charges. When pumping is resorted to, on the other hand, there must be a daily expenditure which, even if small, in the course of a year amounts to the interest on a large sum of money. For example, suppose that the cost for supplies for a small pumping engine was only ten cents per day, not counting in the cost of labor. This would amount to $36.50 a year, which at 5 per cent is the interest on $730. It would be $200 cheaper, therefore, to borrow $500, at 5 per cent, to pay for a gravity supply rather than to pay $30 for a pump which costs ten cents a day to run. This same reasoning may be applied to the cost of different kinds of pumps. One pump may cost $200 more than another, but the saving in fuel and repairs may be sufficient to more than justify this additional cost.

Second, a gravity supply is to be preferred because of its greater reliability. It is hardly possible to imagine any excuse for a gravity supply failing to deliver its predetermined quantity of water regularly day after day. A pumping plant, on the other hand, both breaks down and wears out. Valves are continually requiring to be repacked, nuts drop off and have to be replaced, pieces of the machinery break and require repairs, so that with the best machinery it is almost inevitable that for many days in the year the water-supply is interrupted by some failure of the machinery. In planning water works for cities, an engineer weighs and estimates the value of a continuous service, and even if the gravity supply costs somewhat more than the pumping system, it is in many cases adopted because the greater cost is supposed to be compensated for by the greater reliability of the supply.

Windmills.

Perhaps the cheapest source of power for pumping water is a windmill, and in many cases it proves entirely serviceable. It has two drawbacks which are self-evident. Unless the wind blows, the mill will not work, and, unfortunately, at those times of the year when a large supply of water is most to be desired, that is, during the hot summer months, the wind is particularly light. It is necessary, therefore, when using wind as a source of power, to provide large storage which will tide over the intervals between the times of pumping. Again, the wind may blow frequently enough, but may be so light as not to turn the large vanes necessary to pump rapidly and easily the large amount of water needed. Nothing less than a twelve-foot mill ought to be erected, and, to be efficient, the wind must blow at the rate of twelve to sixteen miles an hour.



A windmill of the best design is made entirely of steel with small angle irons for posts for the tower, and with the mill itself made of galvanized iron. It requires a good foundation and must be well anchored to the masonry piers by strong bolts set well down into the masonry. If the mill is set directly over the well and the storage tank supported on the tower, a very compact arrangement is accomplished and the danger from frost is the only difficulty to be apprehended. However, the tank is often placed in the attic, some distance from the well, to which it is connected by suitable piping.

The location of the windmill requires careful consideration in order that it may receive the prevailing winds in their full force and at the same time be properly located with reference to the well. It must be remembered that the surface of the wheel is exposed to the full fury of a storm, and both the wheel and the tower must be strong enough to withstand such storms. Figure 43 shows windmill and water tank in the vicinity of Ithaca, New York.

Hydraulic rams.

A hydraulic ram is the cheapest method of pumping water, provided that the necessary flow with a sufficient head to do the work is available. It requires about seven times as much water to flow through the ram and be wasted as is pumped, so that if it is desired to pump five hundred gallons a day, the stream must flow at the rate of about thirty-five hundred gallons per day to lift the necessary water.

The two disadvantages of a ram are, first, that a fall of water is not always obtainable or that the stream flow is not always sufficient, and second, that the action of the ram is subject to interruptions on account of the accumulation of air in summer and on account of the formation of ice in winter. In fact, in winter it is necessary to keep a small fire going in the house where the ram is at work in order that this interruption may not take place. Its great advantage is that it requires no attendance, no expense for maintenance, and practically nothing for repairs. It operates continuously when once started, and, except for the occasional interruption on account of air-lock, is always on duty.



Usually the water is led from above the dam or waterfall in a pipe to the ram and flows away after passing through the ram, back into the stream. The water pumped is generally taken from the same stream and is a part of the water used to operate the ram. This is not necessary, however, and double-acting rams are manufactured which will pump a supply of water from a source entirely different from that which operates the ram. The following table from the Rife Hydraulic Engine Manufacturing Co. gives the dimensions and approximate costs of rams suitable for pumping against a head not greater than about thirty feet for each foot of fall available in the drive pipe:—

TABLE XI

======+=======================+=======+=========+===============+ Gallons per Dimensions Size Size Minute -+ -+ - of of required Drive- Delivery to operate Number Height Length Width pipe -pipe Engine + -+ -+ -+ -+ -+ -+ 10 2' 1" 3' 2" 1' 8" 1-1/4" 3/4" 2-1/2 to 6 15 2' 1" 3' 4" 1' 8" 1-1/2" 3/4" 6 to 12 20 2' 3" 3' 8" 1' 9" 2" 1" 8 to 18 25 2' 3" 3' 9" 1' 9" 2-1/2" 1" 11 to 24 30 2' 7" 3' 10" 1' 10" 3" 1-1/4" 15 to 35 40 3' 3" 4' 4" 2' 0" 4" 2" 30 to 75 80 7' 4" 8' 4" 2' 8" 8" 4" 150 to 350 120 8' 9" 8' 4" 2' 8" 12" 5" 375 to 700 120 8' 9" 8' 4" 2' 8" 2-12" 6" 750 to 1400 ======+=======+=======+=======+=======+=========+================+

======================================= Least Feet Price Price of Fall Single- Double- Number Recommended Weight acting acting - - - 10 3 150 $ 50 $ 65 15 3 175 55 70 20 2 225 60 75 25 2 250 66 81 30 2 275 75 90 40 2 600 150 170 80 2 2200 525 575 120 2 3000 750 850 120 2 6000 1500 1700 ===========================

If the length of the discharge pipe is more than a hundred feet, the effect of friction is to reduce the amount of water pumped, but rams will operate successfully against a head of three or four hundred feet. The writer remembers an installation in the northern part of New York State, where two large hydraulic rams furnish the water-supply supply for an entire village, pumping every day several hundred thousand gallons. Figure 44 shows an installation by the Power Specialty Co. of New York, using the fall of some rapids in a brook to pump water into a tank in the attic of a house.



In Fig. 45 are shown two methods of securing a fall for hydraulic rams, recommended by the Niagara Hydraulic Engine Co. The first method shows no drain pipe, but a long drive pipe; while the second method puts the ram in an intermediate position, with considerable lengths of each.

There are other methods of utilizing the fall of a stream, but usually they involve a greater outlay for the construction of a dam and other appurtenances. An old-fashioned bucket water wheel may be used, which, though not efficient, utilizes the power of the stream. The wheel may be belted or geared to a pump directly or may drive a dynamo, the power of which may in turn be transmitted to the pump. The objection to such construction usually is that during the summer the small streams which could be made of service at slight expense run dry or nearly so, while the expense of damming and utilizing a large stream where the water-supply is always sufficient is too great for a single house.

Hot-air engines.

The simplest kind of a pump worked mechanically is the Rider-Ericsson hot-air engine (see Fig. 46), which is made to go by the expansive force of hot air. The fuel used may be wood, coal, kerosene oil, gasolene, or gas, the amount used being very moderate and the daily expense of maintenance very small.



For a number of years the writer used one of these machines to pump water from a tank in his cellar to a tank in the attic, so that running water could be had throughout the house. With an engine and pump costing $100, it was necessary to pump twice a week for about an hour to supply the attic tank and to furnish the necessary water for the family. The following table shows the dimensions, the capacity, and the fuel consumption of the different styles of pumps made by this company:—

TABLE XII

======================================================== Suction and Anthracite Size of Discharge Capacity Cu. Ft. Kerosene Coal Per Cylinder Pipe Per Hour of Gas Per Hour Hour Price - - - - 5" 3/4" 150 gal. 12 1 qt. 4 lb. $ 90 6" 1" 300 gal. 16 2 qt. 4 lb. 130 8" 1-1/4" 500 gal. 20 2 qt. 5 lb. 160 10" 1-1/2" 1000 gal. 50 3 qt. 6 lb. 240 ========================================================

Gas engines for pumping.

During the last few years, on account of the great demand for gas engines for power boats and automobiles, the efficiency and reliability of these engines depending upon the explosive power of the mixture of gas and air has greatly increased. To-day, probably no better device for furnishing a satisfactory source of power in small quantities at a reasonable cost can be found. One engine might readily be used in several capacities, pumping water during the day or at intervals during the day when not needed for running feed cutters; and possibly running a dynamo for electric lights at night. It would be easy to arrange the gas engine so that a shift of a belt would transfer the power of the engine from a dynamo to a pump or to other machinery. In this case the pump is entirely distinct and separate from the engine, and while the gas engine may be directly connected with the pump and bolted to the same bed plate, if the engine is to be used for other purposes than pumping, an intermediate and changeable belt is desirable.

The term "gas engine" is properly restricted to engines literally consuming gas, either illuminating gas or natural gas; but the term is also applied to engines using gasolene as a fuel. The same principle is used in the construction of oil engines where kerosene oil is the fuel instead of gasolene, and it is probable that the latter engines are safer; that is, less subject to dangerous explosion than the former. Whichever fuel is used, the engine may be had in sizes ranging from one half to twenty horsepower and are very satisfactory to use. Any ordinary, intelligent laborer with a little instruction can start and operate them, and except for occasional interruptions they may be depended upon to work regularly. The cost of operation with different fuels may be estimated from the following table, which also shows the cost when coal is used as in an ordinary steam plant, the data being furnished by the Otto Gas Engine Works:—

TABLE XIII

=================================================================== Fuel Consumption Cost of Fuel Per Brake H.-P. Per Brake Fuel Price of Fuel 10 Hours H.-P. 10 Hours - - - Gasolene 10c per gal. 1.25 gal. 12.5c - - - Illuminating gas $1.00 per 1000 180 cu. ft. 18c cu. ft. - - - Natural gas 25c per 1000 130 to 160 cu. ft. 3.25 to 4c cu. ft. - - - Producer gas, anthracite pea coal $4.00 per ton 15 lb. 2.67c - - - Producer gas, charcoal $10.00 per ton 12 lb. 5.35c - - - Bituminous coal, ordinary steam engine $3.00 per ton 80 to 100 lb. 10.7 to 13.4c =============================================================

A photograph of a small (2 H.P.) gas engine made by the Foos Gas Engine Co. with pump complete is shown in Fig. 47. This pump will lift forty gallons of water per minute, with a suction lift up to twenty-five feet, to a height of about seventy-five feet above the pump. The pump gear can be thrown out of connection with the engine, so that the latter can be used for other purposes where power is desired.

Steam pumps.



The use of a steam pump would probably not be considered for a single house unless a small boiler was already installed for other purposes. Not infrequently a boiler is found in connection with a dairy for the purpose of furnishing steam and hot water for washing and sterilizing bottles and cans. Where silage is stored in quantity, a steam boiler and engine are often employed for the heavy work of cutting up fodder. In both these cases it may be a simple matter to connect a small duplex pump with the installed boiler, as is done frequently in creameries, for the sake of pumping the necessary water-supply for the house. Whenever extensive improvements are contemplated, it is well worth while to consider the possibilities of one boiler operating the different kinds of machinery referred to. In Fig. 48 is shown a small pump, made by The Goulds Manufacturing Co., capable of lifting forty-eight gallons of water per minute against a head of a hundred feet. The diameter of piston is four inches and the length of stroke is six inches. It is operated by a belt from a steam engine used for other purposes as well.