line, and the arch ring, if too narrow, will fail like Z' Y', coming down at the crown, even if the abutment is strengthened against yielding by buttresses, as on the left side.

If a metal rib like a truss is employed, the equilibrium curve is still useful. If A B C D is one panel of such a rib, and E F the equilibrium curve for a certain loading, the thrust in the piece D C will equal the thrust in the equilibrium curve at E multiplied by 4 while the thrust in A B will equal that at F multiplied by the ratio, the dotted lines to E and F being perpendiculars to the curve.

A E

AD

CB'

It is not difficult then, if one has a general idea of the location of the equilibrium curve, to determine the weak points in the arch, and to decide at what places the load may with advantage be increased or diminished. It is evident also that the addition of a moving load at X will improve the stability of that road arch, and that an increase of load in the road above W will decrease its stability. If the effects of the steady and moving loads at any point are of the same kind they are added, if of opposite kinds the difference is taken.

Some of the principal points have now been briefly sketched, in order to show why one arch will stand and why another will not. The abutments are supposed to be immovable; if they yield under the horizontal thrust, the arch is endangered, independently of other considerations.

The writer has applied this method of treatment in detail in that volume of his "Graphics" which discusses the arch.

THE DISPOSAL OF SEWAGE ON LAND,

With Particular Reference to Inland Villages and Isolated Buildings where Public Sewers are not Accessible.

GEO. S. PIERSON, M. AM. SOC. C. E., KALAMAZOO, MICH.

Our knowledge of the causes which influence the purification of sewage when applied to the soil has increased markedly during the past few years. It is still an open question, however, whether water which has been contaminated with sewage can be so thoroughly purified as to be entirely safe for culinary

uses.

It is a matter of common knowledge that the purest of natural waters, so far as organic matter is concerned, are those which have undergone prolonged filtration through the soil. And it is well established that the purification of sewage by intermittent application to the soil is the most complete, and where the conditions are favorable, much the most desirable. In fact, it is not unusual to supplement chemical processes by subsequent filtration through the soil whereby the purification is carried to a higher degree.

The dangerous impurities with which sewage is charged consist mainly of different organic compounds in various stages of decomposition. It is not practicable, or in fact desirable, to prevent the decomposition of these organic compounds. With the exception of the small portion which may be consumed as food by animal life, they must be resolved into their elementary substances before they can again be utilized by plant life or otherwise rendered innocuous. It is practicable, however, to so control the conditions of decomposition that it shall not be offensive, and the resulting compounds shall be fixed and rendered harmless by some surrounding medium. The soil is a medium which not only renders these products innocuous but also favors a mechanical separation and aeration very conducive to the rapid disintegration of the putrescible matters contained in the sewage, and, where utilization is desirable, to their absorption as well.

In the report of the select committee on the Metropolis Sewage it is stated that:

"No efficient artificial method has been discovered to purify, for drinking and culinary purposes, water which has once been infected by town sewage. By no known mechanical or chemical means can such water be more than partially cleansed; it is always liable to putrefy again. Processes of filtering and deodorization cannot, therefore, be relied upon to do more than mitigate the evil. Water which appears perfectly pure to the eye is sufficient, under certain conditions, to breed serious epidemics in the population which drinks it. Soils, however, and the roots of growing plants have a great and rapid power of abstracting impurities from sewage water and rendering it again innocuous and free from contamination."

The process of purification of sewage by filtration through the soil is similar to that of burning up or oxidizing the organic matter leaving only a harmless mineral residue, which is soluble and passes off in the effluent, leaving the filtering medium, when properly managed, undiminished in efficiency. The application of sewage intermittently serves to increase the amount of oxidation similarly to opening the draft of a furnace.

The changes produced by wet oxidation, however, are not the same as those produced by heat, there being intermediate processes. In the former the nitrogen of the organic matter first combines with the hydrogen to produce ammonia which, upon uniting with oxygen, produces nitric acid. This, in turn, combines with potash, soda, lime or some other base present in the sewage or in the soil to produce a soluble nitrate. The extent to which this sequence of combinations has proceeded is a measure of the degree of purification of the sewage. The larger the amount of nitrates in the effluent, therefore, and the smaller the amount of ammonia, the more completely has the organic matter of the sewage been destroyed. Later investigations have shown that the earlier of these processes depend on the presence of living organisms.

This has been clearly established by exposing solutions of organic matter, in which the living organisms have been destroyed, freely to air filtered through cotton wool. Under these conditions no nitrification took place. Upon the introduction of a small quantity of soil or other medium containing these living organisms, however, they rapidly multiply and the purification proceeds.

It was formerly supposed that the oxidizing power of a soil depended solely on its porosity, being due to its contained oxygen. Later investigations have shown that a porous medium is not essential and that sewage may, indeed, be nitrified in a glass bottle or when passing over polished pebbles. Although it is proven that porosity is not essential it favors the rapidity of the nitrification in the soil by presenting an immense surface covered with oxidizing organisms and generally well supplied with air requisite for the continued discharge of their functions. In order to more fully illustrate this point the author made the following experiment:

Fifty cubic centimeters of ordinary screened mason's sand, of a fineness of 40 grains per lineal inch, were placed in a chemist's burette, having first been thoroughly freed from moisture by continued drying at a temperature of about 225°. Water was then introduced into the burette from below by aspiration, so as to facilitate the expulsion of contained air until the voids were entirely filled and the amount of water introduced carefully noted. The burette was then opened below and the excess of water over that naturally adhering to the particles of sand was allowed to drain off. From the facts noted the following computations were made:

The total air space in the dry soil was 36 per cent of the cubic contents. The water adhering to the particles of soil was 18 per cent of the cubic contents. The total superficial area of the particles of soil for each cubic foot was 2,200 square feet. The water adhering to the particles of soil for each foot in depth was equivalent to a film of water 1-1000 inches thick and 2,200 square feet in area. Since the purifying agencies within the soil and its contained air have been proved to be active to a depth of at least three feet we may assume that the surface of sewage which is exposed to the action of these purifying agencies is approximately 6,600 times greater when sewage is applied to the soil intermittently than when it is simply impounded over the same area.

The part which these organisms (bacteria) play in the destruction, or rather in the transformation, of the organic impurities of the sewage is stated so clearly and pertinently by T. Mitchell Prudden, M. D., in his book on "Drinking Water and Ice Supplies," that I quote his words as follows:

It has been found that if one starts with an artificial filter bed of perfectly clean sand, containing no bacteria, and floods it with dirty water the water which comes through for the first few days, and for a much longer time if the weather be cold, will be but little, if at all, purified. Its coarser suspended particles may have been caught in the sand pores, and so it may be clearer, but its dissolved organic matter and its bacteria may not be at all diminished. Indeed for some time, strange

as it may appear, the number of the bacteria may have largely increased. In fact, it appears that the pores of such a fresh sand-filter with the organic matter suspended in the water, form a most excellent breeding place for bacteria.

This seems discouraging, but let the experiment go on, and after a while if the dirty water has not been forced through the sand too fast, it will be found that the number of living germs which come out in the water at the bottom is growing steadily smaller and finally the water may be nearly or quite germ free. Now, if the chemist exposes some of the filtered water to his delicate tests he may find that the organic matter which was in solution in the water at the top has already diminished or entirely disappeared, being represented, perhaps, by nitrogen, which has formed harmless combinations with oxygen.

It really seems as if the more of the living, growing bacteria you had in the upper layers of your filter bed, the freer became the water below both in bacteria and organic matter. This is, in fact, the case. We do in this experiment what nature does on a larger scale-make the bacteria fight the organic matter and themselves.

But how is this effect produced? The bacteria are so small that hundreds of them could easily pass abreast through the smallest spaces between the sand particles. What holds them back?

When the sand particles at the upper portion of these filter beds have been carefully examined it has been found that they are, after a few days, completely encased in a slimy gelatine-like envelope, formed of a material which many bacteria secrete around themselves as they grow. This bacteria-formed slime more or less fills the pores of the filter bed, enclosing the bacteria themselves and the sand particles, and catches and holds fast on its sticky surfaces not only suspended matter of various kinds but the new bacteria which come onto the filter and start to work their way down through its pores. Here, many of them, like good prisoners, set to work to make the best of the situation, and if their nature permits, turn to and help to make more of this trap-slime to capture the next comers.

Many of the enlarged germs, however, do not form this material and these may die in large numbers where they lie. On the other hand, this enforced detention is simply paradise for many of the germs. Here they are resting at ease in a slimy confinement, with boundless supplies of just the food they want slowly trickling by them. The food is dead organic matter, which the average bacterium simply dotes on and recks little whether it be in solid form or in solution, so there be enough of it. At it he goes then, and by some wholly inscrutable phase of the life power in his tiny body, asunder fall the atoms which have once been parts of animal or plant. That part which the tiny life spark needs to keep its glow agoing is appropriated. The rest he leaves, its atomic cravings unsatisfied, and only too ready to succumb to the wiles of the ever amorous oxygen, which must always be present in a perfectly acting filter bed.