zers, is then not supported by experiment. On the other hand, the opposite condition is true.

Second, Under-drains leave the land without moisture, and hence in a condition to be affected by a drouth.

This statement seems reasonable, for if the drain works it carries off water, and if the water is gone we have a drouth. I suppose it was not even thought necessary to quote facts to prove such an evident statement. For some reason, the facts are directly at varience with this theory, as I will show you by one or two illustrations. The people with well drained farms have been almost exempt from the evils of drouths, while their neighbors with undrained farms have been literally parched up. These cases have excited common remark in districts where drained and undrained farms were common. This may seem against reason, but is easily explained by the fact that the soil is a great retainer of moisture and that it will retain all that can be of service to the crops, even when thoroughly drained. As showing the amount received and the amount that passes through to the drains, I refer to the following measurements with Lysimeters, in all cases arranged so that none can run off the surface, all water passes through the Lysimeter :

In all the above cases we have the natural conditions as they exist, the Lysimeter is so arranged that the drainage is as nearly perfect as possible, and yet it is seen that compared with the rain fall the drainage is small indeed, and also that it is much less on plats covered with grass than on bare soil. This is due to the fact that the evaporation is no doubt much higher on the plats covered with vegetation than on those without.

Although the average amount discharged by the drains varies from 18 to 36 per cent of the rainfall, yet at times when the rainfall is excessive, the per cent is much higher. The process of drainage is as follows: The water falls on to the earth and is absorbed until the soil becomes thoroughly wet or saturated; at this instant, in case of the underdrains, water commences to flow through them and all excess of water is carried off, yet the soil is still left damp. If there were no underdrains the result would be this, the water would pass into the soil only until it was saturated and literally filled. It then would commence to form in puddles on the surface, and to run off in turbid streams to the larger water courses or lakes. The water from the underdrains

is pure and clear, that from the surface is turbid with the best soil. If the water does not run off through the soil it is filled to the surface and remains saturated with water. This saturated soil is dead, it is impermeable to air and it is unaffected by those agencies of the atmosphere which otherwise would exert a beneficial effect upon it and render it fertile.

This excess of water is only removed by evaporation. Now, evaporation is a process that can only take place through the agency of heat. It requires the beat from one pound of coal to evaporate fourteen pounds of water after it is raised to the boiling point. To raise it from 40° and evaporate it, one pound of coal would only evaporate 12 pounds of water. If an acre of land were covered with water an inch in depth it would weigh 116 tons, and would require as much heat for its evaporation as would be produced by burning 93 tons of coal. No matter whether this heat is obtained directly from the sun or drawn from the ground, it is heat that would have been utilized by vegetation had means been provided for the water to run off. The amount of water to be evaporated is, on a small estimate, six inches, and the heat absorbed by so doing that of the burning of 58 tons of coal, at least. Knowing this, can you wonder that your best undrained fields are cold and frosty? This process of evaporation is bad for your fields, as it leaves the ground with a hard surface almost impermeable to the air. Unless the air can freely enter the soil it has no capacity to withstand a drouth. On the other hand the well drained field is free in wet seasons from excess of moisture. Heat is not absorbed by evaporation, for the water is drawn off through the soil. The insensible channels through which the excess of moisture passes off are in dry times filled with air drawn down and from the atmosphere above, or perhaps upwards through the drains into the cooler and open earth, gives up the moisture with which it is always loaded to the cool particles of soil with which it comes in contact. The result is that the drained and permeable soil is protected alike from excess of moisture or parching drouth. The following table is the amount of moisture claimed to remain in soil after drainage:

So much for reasons. I will now present the experience of a few practical farmers. (Drainage Journal, Dec. 1882): "In 1874, I tile-drained one acre of worthless bog land at an expense of $60. This extraordinary expense was rendered necessary by the natural conditions and surroundings. The work was not completed early enough for ordinary spring crops, so I grazed my lot until mid-summer, when I broke it up and sowed turnip seed. That fall I sold the turnips in the field for $40, or two-thirds of what it cost me to reclaim the land. Since then I have usually grown mixed crops on it, always having satisfactory results. Last year, 1881, the year of dearth and failure, I had it in corn, realizing full sixty bushels to the acre of the very best quality, thus contrasting beautifully with those of my neighbors, whose harvests yielded at the rate of 'sixty acres to the bushel.' This year, when the other climatic extreme has prevailed and short crops are the rule, my tile drained acre is

carrying aloft the golden banner of sixty or seventy-five bushels to the acre. Tile drainage does pay!" W. H. R. in Industrial Times.

The example cited is not an extreme case; it illustrates the common experience of those who have thoroughly drained. It is, however, sufficient to show the value of a drain in dry times.

The third objection of the N. Y. Tribune urges shallow drains as preferable to deep ones. This is not an objection to the drains themselves, and I will only stop here to mention that a large experience has shown that drains between three and four feet deep, everything considered, are preferable to deep or shallow drains. The drain will not be permanent unless at least one foot deeper than the depth of plowing, as the disturbance of the earth will fill the tiles with sediment or displace them. If the drain is very shallow its effects will be essentially local, water will not be taken any great distance from the sides. As the depth of the drain is increased, within certain limits, its ability to drain on each side is also increased. This rule does not go on indefinitely, but so far as I can judge from our experience, and that of others, the best results are with drains about 4 feet deep and not over 60 feet apart. I think that is more economical and fully as good in practical results as where the drains are 3 feet deep and 45 to 50 feet apart. The deep drains also exercise a great influence, which no one can doubt who has had any experience, in making the subsoil more permeable to air and more susceptible to the influences of the atmosphere. The hardest hard-pan will in time become friable and lose its distinctive characteristics.

The fourth objection that the surface wash is increased by underdrainage is so manifestly untrue as to need no attention.

The fifth objection that drainage, together with disappearance of forests, is the principal reason for excessive rains and drouths. This statement is one which can only be proved by long observation. It is held generally that forests tend to make climatic conditions more uniform by acting as a shield for heavy winds and by presenting a large evaporating surface from which supplies of moisture can be drawn during dry weather, it tends to favor the conditions needed for the production of showers. The forests also serve to hold in check the water that would otherwise pass rapidly into streams. It has been claimed that the removal of these forests would tend to the production of seasons of excessive wet or of excessive drouth. In regard to this theory of the effect of forests on rainfall I would say that it is generally admitted, although recently some people are claiming that the facts are a little obstinate and will not array themselves in aid of the theory. This I suppose is so much the worse for the facts.

To prove or disprove that the action of underdrainage tends to aid in bringing about the same result would take a series of minute observations extending over many years. It is true that excess of water is carried off by the underdrains in a much quicker time than it would pass off by evaporation, while it is also true that in passing off it occupies much longer time than where the water all washes over the surface. The amount carried by the underdrains is also somewhat in excess of that taken over the surface. Yet, despite all of this, the operation of the underdrain, so far as it relates to that excess of water which would not pass over the surface is a slow one, for most of this water has to percolate to some depth into the soil and also travel some distance horizontally. By the time this water has reached the outlet the principal surface water, which forms the flood, has been gone for days and the effect of the underdrains so far as the flood is concerned is rather beneficial than otherwise.

With open ditches the result would be different; they would gather surface water and tend to augment the flood. The fact that no doubt caused this accusation is that in 1882 and 1883 there was an almost unprecedented rise of the Ohio river. It was well known that the States on the north side of that river had recently been underdrained. These two things were at once connected as cause and effect. If they really stand in that relation we must of necessity have a flood every year which shall be increased in magnitude as the drains increase in quantity. We had in 1884 instead of a flood a drouth. It has also since been discovered that the Ohio river had been as high at least three different times before underdrainage had been thought of. So that the charge that increased underdrainage was responsible for this state of affairs is to say the least not proved. Honest researches lead us to believe there is in reality no reason to fear any sensible change in climatic conditions due to the construction of underdrains.

In the article of Col. Curtis he calls attention to the destruction of permanent pastures and springs by the construction of underdrains. I make no claim that it is best to underdrain everything. If a permanent pasture is already well balanced as to its water-supply and plant growth, it would be foolish to disturb this relation. Underdrains, however, are in some instances used to advantage, even in such a case. The best piece of permanent pasture that I know has one large drain through it that can be choked up at pleasure. This gives almost absolute control of the moisture retained in the field. I mention this to show that drains are not entirely out of place in a permanent pasture.

Springs. As to the destruction of springs, there is no necessity for this. The surface of the earth can be made tillable, while the spring at the same time can be made available, and it will be found as permanent as before. It is not necessary to have a wet or boggy piece of ground in order that you may have a spring in another place. A drain with cross drains put in judiciously will, no doubt, cut off the spring. This can be opened to the surface in a convenient place, and one of the most perfect of watering places constructed. Such proceedings are common in drainage operations.

2d. The destruction of wells by the drainage of swamps from which these wells depend for a supply, is deplored as a great evil. This very thing should in reality be hailed as a great blessing by the person whose well has its water supply cut off. Put the case in a different way: About one-half mile from your house is a noisome swamp filled with all sorts of decaying matter, dead animals, dead grass, half rotted wood and covered with filthy water in which is a green, slimy, noisome plant. Suppose that this water, by percolation reaches your well and becomes your daily drink. If it is unfit to drink in the pond, it is no better after it has reached your well, except that it may have been strained and may look pure. It may possibly have been chemically purified, but the chances are against that, Your well is a constant danger to you and your family; malaria and typhoid fever lurk in its depths. It may be a convenience, but it threatens your existence, and the removal of the source of contamination from whence it springs is the removal of a temptation certain to prove fatal to you or yours. It is dangerous to drink water from a contamiinated source. Although purification is by no means impossible, it is very improbable.

Again, it is very seldom that the drainage of marshes affect in any way wells in the vicinity. I have no doubt that the cases mentioned by Col. Curtis actually did occur but such cases will seldom be met with. I think that on

the average there would not be one well dried for each one hundred miles of drain constructed. The remaining portion of the article by Col. Curtis calls up objections that have already been answered. Now as to the actual facts in the case. Do drains pay or is it simply that they ought to pay. I call your attention to some statistics gathered by the State authorities of Indiana.

"Two periods of five years were selected, one before drainage was begun, and the other after most of the farms had been drained, the area examined being one township in Johnson county. Before drainage the average yield of wheat for five consecutive years had been 94 bushels per acre; after drainage the same land and tillage gave for an average of five years 194 bushels per The corn crops compared in the same way showed before drainage 31 bushels per acre, after drainage 741 bushels per acre.

acre.

Now for health: The physicians reported during the first period of five years 1,480 cases of malarial disease; during the next period of five years with a fairly good system of drainage there were but 490 cases of such disease.

These results show an increased production of nearly 10 bushels of wheat per acre and of over 40 bushels of corn. This you can readily see would soon pay the cost of drainage, but another thing which is of vastly greater importance, the malarial cases common to badly drained districts had been reduced more than two-thirds. We should expect in a second period of five years to find the malaria nearly gone.

DOES DRAINAGE PAY?

In 1866, the late Chas. K. Carpenter, on his farm at Orion, Mich., had about one acre of swale land just dry enough never to grow a crop, and yet absorbing as much cost in labor and preparation as the rest of the field. This was drained with about 30 rods of drain constructed by opening a ditch 3 feet deep, filling for one foot with loose stones, then covering with marsh hay and straw before filling. The cost was about $10. The next year this piece, with the adjoining four acres, bore a crop of corn that yielded 120 bushels of ears to the acre and was given the first premium in Oakland county. From that day to this, that portion of the field has never failed to produce a good crop. This has paid well, indeed.

Two years later the same gentleman had about one mile of tile laid. This was in the infancy of tiling. The tile used were the horseshoe, without bottom, and they were laid on a board. The man who laid those tiles is still at the business, and no richer, except for children, to this day than then. The tile were laid in, a haphazard sort of a way, two lines working very well and giving good results even to this time. These two lines of tile, about 240 rods in all, have proved very profitable, as they converted some very wet, springy ground into fertile, productive land. The third line, however, was of the kind of drainage that did not pay, [the grade was low and the haphazard style of guessing which way was down hill did not succeed as well as on the other lines. In a short time the drain was choked with soil.

Now, I am of the opinion that the kind of drain that does not pay is invariably of this kind. Such drainage never pays. Failure of drains to operate is usually ascribed to poor tile, or to some general defect, but rarely to the true cause, which is imperfection of grade line and carelessness of laying.

Mr. Homer Ferguson, of Washington, Rush county, gives us the following information in regard to draining some swamp land. (See Drainage Journal,