question, that the witness would address himself also to the other feature of the Mississippi River Commission bill. If you recollect it is stated in the bill that the object of the creation of the Commission is to improve the navigation of the river, or rather devise means for the improvement of the navigation of the river and the prevention of disas trous floods. Now, I understand this bill is to practically carry into effect the latter clause of that bill which seems to have been overlooked as an independent work, and I would like, general, in your statement in reference to the improvement of the navigation of the river, that you would also give us your views on the feasibility of constructing levees of sufficient strength and height to confine the river to its channel and prevent disastrous floods.

The CHAIRMAN. I intended to follow your idea also, but I would like for the general to give us, in the first place, his understanding of levees as a means of improving the low-water navigation of the river and then to follow up with the suggestion made by Mr. Boatner.

Mr. STOCKDALE. Mr. Chairman, first on that point I would like to ask the question as to how long any of this bank protection has been in place in the river and has been a permanent protection to the bank? Mr. BOATNER. With the permission of the chairman I would like to make this suggestion. General Comstock has been for long time in this business and he knows a great deal about the subject, and I anticipate he knows pretty well what we would ascertain, and I ask that he be permitted to make a general statement on the subject to cover the points and afterwards in regard to a matter of detail which any member of the committee would like to know about it could be inquired into specifically.

The CHAIRMAN. I understood that a member of the committee, Mr. Stockdale, desires to ask some question in regard to bank protection; what was your question?

Mr. STOCKDALE. How long has any of this bank improvement been in place, and is it regarded by the Commission as a permanent protec tion when it is so completed?

General COMSTOCK. I think we put in the Bullerton bank protection in about 1882, under water and above water, both. It has stood 8 years very well. In our bank protection we have used a good deal of brush, which has acted under water very well. That which is above low water decays, and as it decays we must add stone so that ultimately when the protection is permanent we would have a layer of stone from low water to the top of the bank.

Mr. STOCKDALE. Will stone stay there?

General COMSTOCK. We have tried it, and we have not lost any yet. From St. Louis down for about 25 miles the river has been improved using that method, and it has stood very well.

Mr. BOATNER. There is work at Vicksburg which was done 8 or 10 years ago which is intact. I noticed it the other day as I came

across.

The CHAIRMAN. Now, general, if you will enter upon the other subject.

General COMSTOCK. In reference to your question as to building levees in improving low-water navigation, I do not think it is a cheap way of improving the river and I do not know that they improve it at all. The trouble is, as I stated before, where the river is excessively wide the water is shallow. At Plum Point Reach there is a place where the river was a mile and a half wide and when we began the work the water went down sometimes to 5 and 6 feet. Now below

that wide place 4 or 5 miles there is a place where the river is narrow and the water is very deep. If you examine the velocities of that deep place and the shallow place I spoke of, you will find at low water the velocity is very slight in the deep place and high relatively in the shallow place. As the river rises the velocity increases in the deep place until it becomes very high. It increases in the shallow place but it never goes as high. The tendency in these deep places of the high velocity is to scour out during floods the material which it takes to one of the lower places, one of those wide places where there is a low velocity and where part of that material is dumped, and the result is during the high water the bed of the river rises in these shallow places. As the river falls again the low-water velocity becomes greater than it is in the deeper places and this falling water cuts out a channel through the sand which has been dumped. This is what the pilots call cutting out bars and is well known on the river, as the river is continually carrying on this process of building up bars; and by building levees a mile and a half apart you are going to increase the velocity so slightly that the question is whether you will increase the velocity enough to more than counterbalance the greater velocity in the deeper water and the greater scouring. That is the theoretical side of it. In 1874 Colonel Suter made a report mentioning the depth of the worst bars of the Mississippi River. We have for 2 or 3 years for part of the river similar data since that time. In 1888 and 1889 it was a very low-water year, so we know what bars there were with less than 10 feet on them then. Since 1880 there have been something like $8,000,000 or $9,000,000 spent on levees below Memphis. If levees had deepened the water it ought to show itself in some degree on those bars, and if the list of bars be examined there are nearly as many in 1888 and 1889 as there were in 1874 and the subsequent years, so that, so far as our experience goes, I do not see-of course this method of the measurement of water is not an exact one, it is only a rough one-any positive evidence that levees have lowered the bars. I am in hopes they will do it somewhat, but it is a thing I am very uncertain about, and therefore I am uncertain that the building of levees is going to materially improve low-water navigation, which is done by increasing the depth over the bad bars.

The CHAIRMAN. I suppose when the river is flooded it flows over the channel works.

General COMSTOCK. At many of them it does.

The CHAIRMAN. What in your judgment will be the effect if levees should be constructed so as to confine the flooded waters of the river within a definite channel; what would be the effect upon low-water navigation if they should be built of sufficient height to control the flooded waters absolutely and hold them within a narrow space, a reasonable space?

General COMSTOCK. As I said, I hoped we could build levees that will produce some improvement in low-water navigation, but I have not evidence enough yet to lead me to think that it is an economical way of improving the condition of the river.

The CHAIRMAN. How much of the river is now leveed?

General COMSTOCK. About 1,300 miles on the river below St. Louis, both banks.

The CHAIRMAN. How much would have to be added to make the levee system complete from Cairo down?

General COMSTOCK. The St. Francis Front has had very little done in many years and the levees are in a bad condition. I do not know

what gaps are in this front, but they will amount to many miles. Then at Helena, or 15 miles below there, there is a gap that will be filled up with the present appropriation within some 30 or 40 miles. Then from Cypress Creek, which is below the Arkansas River, the levees are continuous down to the forts below New Orleans, with the exception of the gap where the Red River comes in, a gap of about 20 miles.

The CHAIRMAN. Is it possible, in your judgment, to construct a system of levees of sufficient height and strength to restrain' the flood waters of the river within a narrow channel?

General COMSTOCK. I am not myself in favor of constructing a system of levees that are narrow. I differ in some respects with other members of the Commission in that they think having levees close together is going to improve the river. I do not feel confidence in it, but they may improve the river. On the right bank of the river, the west side, the levees almost as a rule have followed the bank of the river closely. Where they do that, by any caving of the bank they have to move them back beyond the caving, and the United States has built levees, I think, twice or three times in places on account of the bank caving away. On the Mississippi side the Mississippi engineers have been much wiser, and as a result they have a good many miles of the river leveed 3 or 5 miles back from the river and these levees are perfectly safe. I should suppose that 25 per cent. of the money spent in levees since 1880 has been lost by caving since the levee was built.

The CHAIRMAN. That is 25 per cent. of the money spent on the levees?

General COMSTOCK. Yes, sir; therefore I think levees should be put where they are safe, even at additional cost, and the flood waters going down instead of finding a narrow way find this wide way, which will diminish in some degree the height of the levees ultimately necessary. The CHAIRMAN. You said something about confining the waters within a channel of, say, 3,000 feet――

General COMSTOCK. I do not think I mentioned that. Part of the lower river is hardly more than 3,000 feet wide and the levees are on the banks of the river.

The CHAIRMAN. As an engineer do you think it practical to construct a system of levees of sufficient height and strength to restrain the flood water of the Mississippi?

General COMSTOCK. Oh, yes.

The CHAIRMAN. That cou'd be done?

General COMSTOCK. Oh, yes.

The CHAIRMAN. Now suppose that could be done, would not that restrain the flood waters and scour out the channel?

General COMSTOCK. That is the same question I answered.

The CHAIRMAN. But you say you have some doubt.

General COMSTOCK. I am in doubt about it. I am in hopes it will do it, but our experience on the Mississippi River yet so far is that it has not done it.

The CHAIRMAN. Would this be true, the narrower you brought the levees together the greater would be the probability of scouring the channel?

General COMSTOCK. If I wanted to scour the bed and not to protect the country I should put the levees as close together as I could. The CHAIRMAN. Have you made any estimate of the probable cost of constructing a system of levees which would restrain the flood waters of the Mississippi?

General COMSTOCK. I have not, but this is a general idea, that

H. Mis. 1-70

levees should be made straight and all necks cut off if possible without costing very much more. I have not made an estimate on this, but the Commission submitted an estimate very recently of the cost of raising the levees 5 feet. That estimate was $16,000,000 I think.

Mr. BOATNER. Is it not the experience of the Commission that the river has almost invariably shoaled below crevasses and large breaks in the levees?

General COMSTOCK. It has in a number of cases.

Mr. BOATNER. Is it not the experience of the Commission that these shoals have disappeared on the reconstruction of these levees? General COMSTOCK. In some cases.

Mr. BOATNER. Well, it is the opinion of some of your associates, is it not, that the construction of levees does assist in improving the navi gation of the river in concentrating the current and washing away the accumulations of silt and drift and matter which obstruct the flow of the water and create bars?

General COMSTOCK. Yes, sir. In reference to the last question I might say I entirely agree that scour will occur in all places where the water is tolerably deep, that is, where the velocities are already high, levees will increase the velocity; but the question is whether the scour ing power is going to be sufficiently increased in the wide, shallow places. They are the places which obstruct navigation.

Mr. BOATNER. I understand you to say that the velocity being so high in the deep places causes a scouring and causes a corresponding deposit in these wide reaches of the river?

General COMSTOCK. There is a deposit in the wide reaches.

Mr. BOATNER. And when the water fell this deposit would be scoured away by the force of the low-water current?

General COMSTOCK. In the shoal places.

Mr. BOATNER. And if the levees are allowed to give way at these wide places, and in addition to the slackening of the current to which you have referred, so as to cause the eddies and swerves which invariably follow these crevasses, would not these breaks cause a much larger deposit than would be precipitated if the levee should be maintained! General COMSTOCK. I would say, if these views I hold are correct, if you had levees built so as to hold all the water where the river was very wide, and you were to take these levees away absolutely for 20 or 30 miles that the depth on the worst shoal would not change very much. Mr. BOATNER. That is if all the levees were away?

General COMSTOCK. Take Plum Point reach, for instance. If this bar-building process is true which I have described, suppose with the original great width you removed all the levees for 30 miles, that bar would still remain there; it would not change much.

Mr. BOATNER. You do not think it would increase the bar and lower the level of the river to allow the water to escape from the bank at a shallow place?

General COMSTOCK. Now, if you make a small crevasse, say about 1,000 feet wide where the river was deep, it might be there you would find some shoaling; but if you take a reach of say 20 miles, I think the bar would not be much worse after taking the levees away than before, or be much better after building the levees than before.

Mr. BOATNER. You are of the opinion, however, that it is entirely feasible and practicable to construct a line of levee on either side of the river which will prevent disastrous floods.

General COMSTOCK. Oh, yes; reservoir embankments are built 40

and 50 feet high, and this is merely a reservoir embankment to endure a short time, 50, 60, or 100 days.

Mr. STOCKDALE. About what difference would be the distance between the levees following the river and building them across the points as you suggest?

General COMSTOCK. It varies a great deal at different parts of the river. At many places of the river the length straight down the river is half of what following the river would be.

Mr. STOCKDALE. And how much higher would the levees have to be, say those on the banks, taking into consideration the fall of the land as it recedes from the river on the one hand and the closer you get to the water on the other?

General COMSTOCK. We have not had surveys made over many of these points, and as to that question I could not answer positively. The river has wandered so much through the valley, and there are so many ridges, that it would be difficult to tell without any accurate survey. The shortening of the levee about counterbalances the increased height for some of the points on the St. Francis front, but whether that would be true on the whole or not I am unable to say.

The CHAIRMAN. Suppose you were to construct levees simply to prevent disastrous floods, where would you build them?

General COMSTOCK. It is difficult to say where, because localities differ a great deal. Wherever I could cut off and cross a neck without increasing the expense I would. Wherever I could build it back from a caving point without increasing the expense largely I would build it away back, so as to avoid the difficulty of building a high levee every 3, 5, or 10 years. If you wish to have a permanent and complete system of levees and to take care of them when close to the river you have to protect your cav ng banks. That is done on the Po. The bank must be protected by stone and brush work just to hold these levees. The CHAIRMAN. To protect the bank from caving.

General COMSTOCK. Yes, sir; with a view to the safely of the levees. One-third is probably too large, but there are long reaches protected, and you find levees 15 or 20 feet high protected by stone.

Mr. BRICKNER. For instance you are speaking of building these spurs like that [illustrating] and you say it will increase the velocity of the water.

General COMSTOCK. Yes, it of course diminishes the water way, and anything that does that increases the velocity.

Mr. BRICKNER. Will not it have the tendency to throw the water on the other side?

General COMSTOCK. Yes, sir; that is the reason we have to protect the opposite bank in order to prevent scouring.

Mr. BRICKNER. Now if you had a continuous levee would not that increase the velocity of the water without throwing it on the other side of the stream?

General COMSTOCK. If we had high levees that would increase somewhat the velocity of the water.

Mr. BRICKNER. After you have increased the velocity of the water would not that have a tendency to scour out the channel and make it permanent?

General COMSTOCK. If the only mission of the rising water was to scour, it would. The velocity is increased everywhere as water rises. Now, it might be supposed as the result of that, that everywhere the bed would be scoured out when the water height was increased 5 or 10 feet; but the river in fact does not scour out every where. During high