the connections are best made at the joints, by breaking off a portion of the side of each piece of tile which is to receive the incoming drain Where horseshoe tile are used the connection is made at the center of a piece of the receiving tile, as shown in Fig. 71, where a is the tile of receiving drain, and 6 the tile of the incoming drain. When tile are laid by commencing at the upper end of the drain, the upper end of the first tile laid should rest firmly with its entire end against a brickbat, or other close-fitting surface, so as to prevent the ingress of sand or mud at a point where it is not likely to have a sufficient current of water to carry it off. It is always best to surround the points of junction between the drains by small stones, and these covered with straw or turf, so as to prevent the introduction of sand, silt or mud.

FIG. 71.

CHAPTER XIII.

OBSTRUCTIONS IN DRAINS.

The Central Society of Agriculture, at Paris, having investigated the causes of the obstructions in pipe drains, Mr. Barral said that there are three causes, viz : deposits of carbonate of lime, sediments of hydrate of peroxyd of iron, and intrusion of roots.

It was remarked that, in general, those obstructions had a primordial cause in the defective laying of the drains: some of them lacked sufficient declivity; others had pipes with imperfect joints; but the most frequent occurrence was the intervention of roots. However, a fall of 1-500 is sufficient to carry away the roots, and these are found in balls at the exit of the main drain.

Obstructions are easily detected by an extraordinary moisture which is manifested in the soil at the place where the pipe is obstructed.

Mr. Hervé Mangon, Drainage Engineer of the French government, says:

"I have found obstructions caused by sediments of carbonate of lime and oxyd of iron. I will present the result of my studies, on these two classes of deposits, and indicate the means by which I prevent them.

"Calcareous Obstructions.-Spring water sometimes contains carbonate of lime in sufficient quantity to produce incrustations; that is, it deposits calcareous salt; the same phenomenon takes place within drain pipes, the section of which rapidly decreases, until it does not allow any passage for water, and soon the profitable,

wholesome effect of drainage, which is established at great expense, is entirely

lost.

"Water, thus impregnated with carbonate of lime, does not dissolve it, unless it is acted upon by carbonic acid gas, which it also contains; water remains limpid as long as that gas is not disengaged. The calcareous deposit is produced only when the quantity of this gas is no longer in proportion with the calcareous salt present in the water.

"In order to prevent the formation of the calcareous obstructions in drain pipes, all that is required is, to prevent the separation of the carbonic acid gas from the water which flows through the pipes. This may be easily accomplished by protecting the water in the pipes from communication with external air.

"The atmosphere which is confined within the subterranean ducts soon becomes impregnated with a full proportion of carbonic acid gas, as compared with the volume that is dissolved in water; this latter gas does not then any more tend to disengage itself; water charged with its calcareous salt, preserves its limpidity; and it may flow forever without impediment.

"This theory is very readily put into practice. A pneumatic pipe, set upright a few yards above the exit, and others, if necessary, at the point of junction of the most important main pipe drains, will be sufficient. These pneumatic pipes are made of two or three large pipes, well joined together, laid over a flat stone, and covered with another. Some mason work ought to be laid around and beneath the upright, and the horizontal pipes connecting with it; those flowing in must be placed a shade lower than those flowing out; water will thus intercept the air, and the object is attained; that is, carbonic acid gas will be retained.

"Ferruginous Obstructions are formed by sediments more or less impregnated with oxyd of iron, and may be of a red, dark brown, or pale yellow color. When precipitation takes place in quiet water, there appear on its surface rainbow-like cuticles, which are sunk at the bottom by the slightest motion of the liquid. That sediment soon obstructs the pipes and completely stops the drain.

"Waters containing such deposits are met with, especially, in soils strongly impregnated with either oxyd or sulphuret of iron, in marshes, turf, and lands which are exposed to filtrations from woods situated on a higher level. The acids named crenic, and apocrenic, also perform an important part in the formation of the above deposits. The elements of the soil have, of course, a great influence in the case; most of the deposits contain large quantities of clay, sand, or detritus of vegetables; so that all the analyses presented widely different results.”

Without following the author in the minute chemical demonstrations, we proceed with a practical and very interesting experiment. He says:

Having collected a fresh deposit, with the very water in which it was formed, I put it on a filter and obtained a perfectly clear liquid; which, being placed into flagons entirely full, and well corked, or within an atmosphere deprived of oxygen, remains transparent. Having exposed one flagon to the action of pure oxygen and another to the open air, both became dim after a short time, and allowed the ochre-like substance, which is the basis of the aforesaid obstructions, to settle or precipitate.

"This substance, which is the same that settles in the drains, was easily separated from the liquid; being exposed to the air, it became more and more reddish, until, after a few hours, no further change took place; being then inclosed within an air-tight flagon, it soon resumed a dark brown and almost black color. After a few weeks, the same sediment being placed again on the filter, the result was the same; that is, a clear liquid, that became dim by contact with air, and deposited the identical yellow substance. On the other hand, the matter left on the filter resumed the reddish tint which it possessed when placed in the flagon."

The same operation may be repeated any number of times on the same sample, with the same result.

It is then evident that this body presents the double quality of becoming insoluble by its oxydation, and of reducing itself, when left alone, so as to become partly soluble.

The above may be summed up in the following two propositions: First, the water which causes ferruginous obstructions within drain pipes, preserves its limpidity, and gives no sediment when not in contact with the oxygen of the air; second, the deposit recently formed may exercise a reducing action upon itself, which causes it to resume in a great part its soluble property.

From these two facts, it is easy to conclude that pneumatic upright pipes, as described above, will prevent the formation of ferruginous obstructions by excluding the oxygen of the air, as well as calcareous sediments, by including carbonic acid. gas.

Brandt had observed that the water impregnated with ferruginous matter collected from the bottom of a meadow, kept in open bottles, began to thicken at the end of three days, and to deposit flakes after five days. The occurrence in some experimental holes made in the meadow produced similar results. As the drain water, even under the most unfavorable circumstances, does not admit of so long a stay in the pipes, Brandt, for the sake of further observation, made an experiment in which he employed three tubes of 120 feet in length, 3 feet in depth, and 6 inches fall (on the whole length,) to be laid in the meadow in question.

The tube A was provided with a wooden discharge pipe two feet long, which

was perforated in an oblique line, and placed so that the discharged water was compelled to fill the opening of the tube.

The tube B had a free discharge pipe.

The tube C had likewise a wooden discharge pipe, which for a length of 5 feet was stamped around with clay, in order to produce a damming of the water in the tube.

The works were undertaken in December, 1852; eight days, however, after the three tubes had been laid, all the drain water was turbid, the openings assumed an orange color, and a short time after, when it rained, the tubes dischargedowing to the more violent intrusion of bottom water-a large amount of oxyd of iron. After a minute investigation, the cause of this occurrence was found in the fact that the single tubes were not placed in the ground in a mathematical straight line; but that they, deviating from the latter more or less, had here and there some points of stoppage in which the water remained stationary, and the formation of oxyd of iron took place slowly, but uninterruptedly. Stronger water currents in the tubes overcame these stopping points, and carried away the sedimentary

matter.

The tubes A and B were obstructed in May, 1853; the third, C, was constantly kept clear by the frequent damming of its own water, effectuated by closing the discharge pipe with a tenon. In order to see how high the water was dammed in the tubes, the tenon was perforated, and a small glass tube placed in the perforation. Two or three days were generally sufficient to press the water to the margin of the small pipe. After the removal of the tenon, the water, filling the entire space of the pipe, flowed off with the deposed substances of iron, and it did so, finally, in general very pure; which result justified the opinion that in this manner an arrangement had been found for protecting against obstructions from oxyd of iron. The draining of the meadows, undertaken in the fall of 1853 and spring of 1854, was then executed by tubes or pipes 20 perches long, laid at the upper end 24 feet, on the lower 3 feet deep. The tubes were laid with great care, and clay slightly stamped around; the discharge pipes were of wood, and led into a ditch, which latter could, by means of a dam, in two days be dammed up 1 foot above the highest point of the drain pipes. By alternate damming and discharg ing, repeated every fortnight, the drain tubes had, up to the middle of 1855, remained free from any obstruction.

Tischendorff tries to remove the obstructions occurring in the drain pipes by pressing water into the tubes at the upper end of the obstructed pipes by means of a simple pump-work. (Zeitschr. f. d. Landw., 1855, 64.)

There are no definite reports on the success of the funnel pipes recommended for the prevention of intrusion of quicksand. (cf. Jahresb., 1854, I, 69.)

Dr. Motherby-Arensberg (East Prussia), reports that in draining in quicksand he had left none of the means recommended untried, but found none always reliable, and that he now gives preference to the following plain method, the principle of which consists in as speedy a performance of the successive operations as possible, in order to prevent the movement of the quicksand. The contemplated ditch is first thrown out deep enough to allow only one more cut to the stratum of quicksand; into the walls of the yet shallow ditch leveling pegs are driven sideways, and to them is fastened a cord, by which the depth can at any instant be correctly ascertained-this being the most important item in the rapid succession of operation. The workmen now begin one after the other, and so close to each other that the necessary free movement only is allowed to each. The second workman commences only after the first one has made his first cuts; the rest proceed in the same way, so that they stand in their work entirely by steps, and the last must constantly be prepared with his hook, ready to receive the tiles and place them accurately and quickly, so that they may be immediately covered by a workman stepping over the ditch, with one foot of earth. In order to be perfectly sure as to the work being everywhere done right, stoppages are made from time to time, which, if arrested, furnish the best proof whether the work has been perfectly made, or where the mistake is which as yet can easily be remedied. In order to make these stoppages, the drain ditch is closed from distance to distance by a small loam dam; the pipe itself projecting from this dam is closed by a cork; the water is then permitted to gather in order to observe whether after removing the cork, a complete discharge of water takes place:

As to the intrusion of roots, Mr. B. de Latour states that a pipe drain, four feet below the surface, being choked up, he ordered it to be repaired; that a great number of thread-like beet roots, ten to twelve feet long, had penetrated and filled the largest pipes; that in another field carrots had caused the same accident; that potatoes had not done it, and he feared nothing from the roots of fruit trees and vineyards.

Mr. L. Giraud and Mr. Th. Galos, from the neighborhood of Bordeaux, state that pipe drains, in the vineyards of that district, are protected against the intrusion of roots, by surrounding the pipes with straw, after having covered the joints with short pipes or collars.

CONCLUSION.

We have now discussed all the prominent principles involved in underdraining, and have given such practical directions for determining the construction of the