in constructing them would have been taken from the remainder of the levee line, which would have been necessarily weakened thereby, and crevasses would therefore have been much more frequent.

In fact, if it could be predicted that the next great flood would be similar to the last, even a somewhat lower provisional grade line would be desirable in certain portions of the river, as 586 miles of levees have not been constructed to this grade, and some 53,000,000 cubic yards must be placed in them to create the cross section which has been adopted by the commission. But no two floods are similar. The grade line established by this flood will be subject to material changes, arising from variations in the discharge of the White, Arkansas, and Red Rivers, or even from local rains.

There is appended a table which gives the heights attained by the river at various localities during the last flood, the previous highest waters, the provisional levee grade, and the estimated high water during the flood of 1912 if no crevasses had occurred. It will surprise many to learn that at none of the stations in the table. the flood of 1912 reached a height equal to that of the provisional grade line, nor did a crevasse occur in any levee that was built to the grade and given the cross section established by the commission, except possibly at Hymelia.

If the recommendations of the commission, made some 15 years ago, had been carried out, this disaster, to a large extent, would have been averted. I do not mean to imply by this statement that the provisional grade adopted by the commission is the ultimate grade to which levees should be constructed; in fact, they must ultimately be built at least from 2 to 3 feet higher; but that if the provisional grade and cross-section had existed throughout the valley, wherever the flood attained a height greater than the provisional grade, there would have been a good fighting chance to hold the levees by topping, while with defective foundations and weak section, the battle was lost before the river could attain that height.

As a result of this flood the commission does not recommend any immediate change in its provisional grade; on the contrary, it is of the opinion that the first work to be done is to strengthen the foundations wherever any weakness has been observed, then to bring the section to standard dimensions. When the levee line is uniformly perfected to the provisional grade, its further enlargement will be advisable. Excessive strength in one locality with the necessary undue weakness at others should be avoided.

CAVING BANKS.

While about 2,500,000 cubic yards of the levee line were destroyed by crevasses during the last flood, over 4,300,000 cubic yards had to be abandoned during the past year on account of caving banks. The loss from crevasses is considered a national calamity, while that from caving banks is scarcely noticed. But I desire to particularly invite attention to the drain upon the community this caving of levees into the river has become. It requires an expenditure of nearly $1,000,000 annually to replace them. The Mississippi River Commission appreciates the relief that Congress has afforded them by its proviso that $4,000,000 of the $6,000,000 appropriated by the last rivers and harbors bill must be expended on levees. It precludes the

use of any funds for the protection of city parks or even city fronts. But there is a danger from too close a limitation of the powers of the commission. It frequently is cheaper to construct a bank revetment than to rebuild a levee which is caving into the river. I apprehend that under the present act several hundred thousand dollars will be wasted. Because of its limitations levees must be constructed where bank revetments are more desirable.

FOUNDATIONS.

The advice which the commission has received on the use of concrete, steel piles, triple-lap sheet piling, and other patent inventions for levee construction, would fill a large volume. I will not detain you with a discussion of these devices further than to state that we are convinced from the results of the late flood that greater care must be exercised in securing the levee foundations, but whether this result will be attained by an enlarged muck ditch, a wall of concrete or sheet piling, or other means, is dependent so much on local conditions that no general plan can at present be formulated.

CONCLUSION.

The flood of 1912 affords no argument for the abandonment of levee construction. It has simply attained the height which Gen. Comstock and Maj. Starling predicted the flood of 1882 would have attained if the river had then been confined. It has cleared the atmosphere of certain false theories, and we can now resume operations with a definite knowledge of the problem before us. We are passing through the same experience European nations have had. Levees have been tested for ages and have proved uniformly successful when built of adequate dimensions. During the progress of construction there were disasters on foreign rivers as well as in the United States. No other method of relief from floods has been successfully applied to large streams.

Originality is a very desirable quality in an engineer, but there is danger of confusing originality and ignorance. When a proposition with which he is unfamiliar is presented to him it is his duty to follow the instructions placed at some railway crossings, to stop, look, and listen. He should investigate what has been done in the past, and seek to discover if there is no precedent for his action.

It was said several thousand years ago that there is nothing new under the sun. The saying is true to-day. To adopt a project, even though popular, that has been tried, found wanting, and rejected by our forefathers, is not progress, but retrogression.

Table of gauge readings of flood heights and provisional grades, Mississippi River, Cairo to Fort Jackson.

THE PROBLEM OF THE MISSISSIPPI RIVER.

The great flood of the Mississippi River of last year-the largest in recorded history-when the levees were overtopped or carried away bodily, and vast areas of the valley were inundated, has created a doubt in the minds of the public as to whether the method of control by revetment and construction of levees was not a failure. This doubt has been freely expressed in the many letters which have been published during the past year in the columns of the Scientific American. We have made no comment upon these letters, many of which suggested alternative and supposedly better plans for the control of the river, and our silence has been due to the fact that we were making a study of the problem from every possible source of information, with a view to determining for ourselves whether the present plans for the control of the river, or some other, were the best to apply in grappling with and controlling this stupendous problem.

We have come to the conclusion that the present plan of the Army Engineers of protecting the banks of the river by revetment and raising the banks by artificial levees to a sufficient height to prevent overflow is not only the best way to control the river, but the only

way.

If it be asked whether the disastrous inundation of last year does not spell failure, we answer emphatically, "No." The inundation occurred, not because the plan was faulty, but because it was incomplete. It was also due to the fact that the existing levees were built only to a sufficient elevation to control the highest flood on record, which the flood of last year greatly exceeded-the maximum flow reaching the enormous total of 2,300,000 feet per second, or 12 times the amount of water that passes over Niagara Falls.

The trouble with the Mississippi work is not that the plans are wrong, but that they have been carried out piecemeal, and in a somewhat happy-go-lucky manner. The Nation should apply to this great work the lesson which it has learned at Panama. A new grade line for the summit of the levees should be established, said line being well above the height reached by the flood of last year; a liberal estimate should be made of the total cost of building these levees, and of protecting the adjacent banks of the river throughout the whole length of the levees with revetments; an estimate should be made of the largest annual appropriation of money that could be efficiently expended by the largest force that could be concentrated upon the work; and finally the execution of the work should be placed entirely in the hands of the Army Engineers with a Col. Goethals in supreme and unhampered control.

Such an estimate of the total cost of a completely leveed and revetted Mississippi River has been made by the Army engineers under the Mississippi River Commission. The total expenditure would be

about $70,000,000 for the levee work and about $90,000,000 for the revetment.

Is complete control of the Mississippi River and the absolute prevention of disastrous floods worth the expenditure of $160,000,000? The Scientific American is decidedly of the opinion that the money would be well spent. In the first place, the completion of this work would afford protection to 29,000 square miles of land. The increased value of the land, due to protection, is shown by a statement of Col. Townsend, president of the Mississippi River Commission, who has recently testified before the Committee on Rivers and Harbors in the House of Representatives that 20 years ago, when he was first stationed in the St. Francis Basin, land in that vicinity could be bought for a dollar or two an acre, whereas to-day it is worth anywhere from $20 to $50 and even $100 an acre. Furthermore, there is the humanitarian consideration that this work would prevent the great loss of life and destruction of property which occurs when the river breaks loose. And, finally, there is the consideration that the completion of this task will constitute a great national work of engineering comparable, in its magnitude and beneficent results, with the execution of the Panama Canal.

We will now proceed to discuss the criticisms of the present plan and the suggestions of alternative schemes of control which have been made in the many letters referred to above. It has been stated that the whole principle of levee building is wrong; and this for the reason that the matter brought down in suspension is deposited along the bed of the river, which is continually being raised, that this necessitates a raising of the levees, which must go on indefinitely. As a matter of fact, what takes place is this: When the floods come down, the deep pools are scoured out and the material is deposited on the shoals farther down the river, causing a temporary raising of the bottom at these points. As the river falls, the action is reversed, the bars are scoured out, and the sand is deposited in the next pool. Careful surveys for several decades show that not only has there been no raising of the river bed, but the cross section of the river has slightly increased.

As to the proposal to control the Mississippi by building vast reservoirs near the headwaters of the river and its tributaries, it may be said at once that the magnitude and cost of such reservoirs and the enormous areas of land that would have to be condemned, render such a scheme impracticable. Its advocates have failed to realize the stupendous magnitude of a problem which involves the control of flood waters that sweep down the Mississippi River at the rate of 2,300,000,000 cubic feet per second. Testifying on the point, Col. Townsend said before the House Committee: "If you were to destroy the whole State of Minnesota-that is, stop every bit of water flowing over it-it would not have made a difference of three-tenths of a foot in the height of the last flood at Cairo." Again, if, as has been suggested, the St. Francis Basin were converted into a storage reservoir and the floods were thereby reduced 3 or 4 feet in height, it would be necessary to sacrifice no less than 7,000 square miles of country, or the area of a good-sized State.

Another favorite scheme contemplates the diversion of the Mississipp or of a large portion of its flood waters, by means of subchannels, or canals," excavated on one side or the other of the river. This