Many a stream to-day in forested areas there does not flood the adjacent bottom lands, while a neighboring stream, with approximately the same rainfall and under the same conditions, except that it has been deforested and is largely cultivated, does, with every little rain, get out of its banks and flood the lands on either side. So true is this that in many cases the people have practically abandoned those lands. Over great areas they are practically worthless for agricultural purposes.

I remember that on my father's farm, when I was a small boy, the most valuable lands we had (and this was commonly true over large areas in the South) were the bottom lands. The stream was perhaps a rod wide and the banks were 8 or 10 feet high. The water level was 6 or 8 feet down below the general bottom land, or flood-plain level. To-day if you go to that farm you will find the stream channel filled with sand. The lands that were formerly worth a hundred dollars an acre are practically worthless; many of them are growing up in willows and bushes, because now every little bit of rain that comes floods the whole land, drowns out any crops that may be planted, and spreads sand over the land so that not even grass will grow on it. You can not turn it into meadows; it is practically worthless; and it is due to this change the erosion of the steep slopes, bringing in materials that fill up the channels and convert what were efficient means of rapidly draining away the flood waters into very sluggish, inefficient means of draining off these same flood waters. The flood waters accumulate, rise higher than formerly, spread over the lands oftener, and cause much greater destruction.

Mr. LEVER. Is that condition general, do you say?

Mr. GLENN. That condition is a very general one. It obtains over some thousands of square miles-and not in one section alone, but up and down the length and breadth of the Appalachian Mountains and the adjacent Piedmont Plateau.

The CHAIRMAN. What particular section do you refer to as having lived in?

Mr. GLENN. When I spoke of my boyhood home, it was in the Piedmont section of North Carolina, at the foot of the mountains, on the east, not far from Kings Mountain. That condition obtains in Alabama, across Georgia, across upper South Carolina, North Carolina, and across Virginia; and it also obtains within the mountain area and west of the mountain area, across Tennessee and up into Virginia. The filling of the stream channels that formerly were deep and were efficient avenues for the rapid discharge of flood waters has practically destroyed their efficiency for that purpose; and rains that formerly would not have caused a flood at all now do cause floods, and cause higher ones than formerly. So that Professor Moore's statement is directly opposed to the facts that obtain down there over thousands of square miles, and could not have been made by a man familiar with actual conditions in the field.

Mr. LEVER. Do you know anything about Professor Moore's scientific training?

Mr. GLENN. I do not, and would not be able to throw any light at all on the reasons for his conclusions. I may say this much in general: His conclusions are doubtless perfectly honest ones, but I believe they have been reached by considering imperfect data. He did not have the data that he should have had to draw those con

clusions from. Had he spent even a few weeks in the southern Appalachians, examining actual conditions in the field, he could not have drawn the conclusions that he did. They are undoubtedly erroneous, as any man could at once see who had been in that region for any length of time-not even giving it special study, as I did, for four years.

Mr. COLE. Did you ever hear Professor Moore's competency as a witness on this proposition objected to prior to his adverse decision? Mr. GLENN. I have never heard him spoken of in connection with it at all prior to this.

Mr. LEVER. He is a new Richmond in the field, is he not?
Mr. GLENN. That was my impression.

There is another feature connected with the matter of floods. Deforestation undoubtedly decreases the low-water discharge of the streams. If you measured the total annual discharge from a forested area, and then chopped off the trees from the entire area and again measured the total annual discharge, I do not know whether there would be much difference or not. That is not at all a vital question, to my mind. It might be that there would be considerable difference. I do not know. I am not really greatly concerned with it. But I do know this, as the result of my experience of four years down there:

In the forested area the discharge is regulated. Doubtless you have heard that time and time again; but I want to repeat it as a result of direct study of that particular point down there, that forests do regulate the flow of the streams that drain them, and that during floods the discharge from forested areas is not as great as it is from unforested areas. With a heavy rain the streams of a forested basin will rise slowly; they will stay high for several days, and will go down slowly. The streams of an unforested area will rise very rapidly, become raging torrents, and run down rapidly.

In that connection I should like to call your attention to a fundamental oversight in one of the conclusions that Professor Moore has reached. His statements are utterly inconsistent with each other. He says, finally :

Floods are not of greater frequency

I feel perfectly confident that they are, but this is the point-
And longer duration than formerly.

Taking one of his other conclusions in connection with this onethat as to their not being higher-it is inconsistent that floods should be both higher and longer in duration from similar rainfall. In other words, the higher a flood is, the more rapidly it will run down. So that if the height of a flood is increased, it is because more water is delivered there at once; and that water goes off as a great tide, sweeps rapidly past, and as rapidly as it rose it falls again. An unforested area flood is one that passes rapidly. A forested area flood is one that passes slowly, because the water is fed in slowly to supply the flood and does not reach enormous heights. In proportion as you increase the height, you increase the slope of the front part of the flood wave that is going down the river slope, and consequently increase the velocity with which it goes. It rapidly runs itself out, in other words.

If you attack the problem along any line and I have tried to work it out in as many ways as I in my ingenuity could suggest the inevitable conclusion is that the higher the flood crest from the same amount of rainfall, the shorter lived it is going to be; the lower the flood, the longer lived it is going to be. I have not the time to go into half a dozen other points of approach to the same problem; but I can find no avenue of approach except one that reaches that conclusion. And I can not see how the last conclusion-that the floods are longer can be sustained by any argument that could possibly be devised; certainly by no observation in the field.

Let us take next the question of stream filling. We must look on a stream just as you would on a railroad. A railroad with a certain equipment is able to carry a certain amount of freight. If you deliver to it more material than its capacity for moving, it is bound to become congested; it is bound to let freight accumulate because it can not possibly carry it away. A stream with a certain amount of water and a certain velocity can carry a certain quantity of sand and gravel. Nature, through long, long ages, fitted the cross sections of the streams, their slope, etc., to carry away efficiently the material furnished to them under natural conditions. As a result of that we find that the mountain streams, under natural, normal, forested conditions have cleared out deep rocky channels.

That photograph before the window there of the gorge of the Linville River, in the western part of North Carolina, shows a natural channel, with a stream flowing and scouring on a rock bottom, with rock sides. If man steps in and by deforestation and erosion furnishes a much greater quantity of sand to that stream, the time will come when, with the increasing quantity of sand, the stream is no longer able to carry it all away. The stream is then bound to let it accumulate there. You have really overloaded the stream. It can not handle the stuff you are giving to it any more than a railroad can always handle all of the stuff that you give to it. It accumulates along the stream upon chance obstructions, such as bowlders, behind logs, in bends, and anywhere that it can lodge. It accumulates more and more, until finally the rocks are entirely covered, and the stream no longer flows on the rocks, but flows within banks of sand and gravel, etc.

Alluding to the source of the sediment in the Mississippi, about which the chairman was asking a while ago, I will state that the material that you would find anywhere in the Mississippi flood plain is material that has come from many diverse sources originally far back, much of it, in the mountains or the upper reaches of the waters. A given grain of sand has traveled a few miles with this flood and lodged, and then it has been carried down a few miles with the next one and lodged, and a few miles with the next one, and ultimately it works its way out to the Gulf. So that the immediate cutting of banks and building of bars a few hundred yards below is merely incidental a step in the general progress of that material in that interrupted way from the headwaters down to the Gulf ultimately. Not all of it comes from the headwaters, it is true; because if you take the total area of the Mississippi drainage basin, the uplands all over it in cultivated fields are being eroded to a certain extent; they are furnishing sand and gravel and clay to the stream, so that it comes

31735-VOL 3-10-3

from all parts of the basin; but the more active zone is the headwaters.

The CHAIRMAN. Is it not true that the headwaters of the Mississippi, including the headwaters of the Missouri, which is really the greater stream-is it not true that they are in soil of a character that does not easily_erode?

Mr. GLENN. I am not personally familiar with the headwaters of the Missouri.

The CHAIRMAN. And is it not true also that these headwaters are in a region of very limited rainfall?

Mr. GLENN. Much of that as to rainfall is undoubtedly true in the case of the Missouri. I am not personally familiar with conditions there, but I do know, from various sources, that enormous volumes of sand comes down the Missouri. Whether they come from the headwaters or are picked up by the streams as they cross the plains I do not know, but I do know that in the southern Appalachians there is an enormous mantle of disintegrated rock material standing on very steep slopes in such a way that the moment you remove the forest cover it begins rolling down into the streams and working its way down, little by little, into the larger and larger channels.

In that connection I wish to call your attention particularly to the results of examining a number of the navigable streams that head up in those mountains. There were two basins especially that are contrasted-the Tennessee and the Coosa, Alabama-both heading in the southern Appalachians. In the Tennessee basin very much of the mountain headwater area is deforested. The sand and gravel that works down that stream has worked into the navigable portion, and is to-day accumulating there behind any chance obstruction wherever it can possibly find lodgment. It is building islands. There are a great many islands in the Tennessee, and those islands are growingsometimes growing rapidly. Islands that a few years ago were just above low water are to-day, in some cases, cultivated in corn, they have built up so rapidly. The growth of an island of that kind is bound to displace the channel of the stream, and the steamboat channel that once ran in a certain course now curves around somewhere else. They find in the middle and lower parts of the river that at such a point they must be perpetually rearranging the lights by which the boats keep the channel and keep their bearings. We find that at such places the banks are caving and engulfing houses and orchards and lands as the result of the accumulation of material in the channel, deflecting it and causing undercutting of the banks. Where dikes are built for the sake of improvement, very soon immense quantities of sand and gravel lodge behind them and build them up, growing islands back of them, as it were. A natural bar that does not show above low water even we find increasing in the quantity of gravel on it, so that a channel dredged by army engineers this year across one of those bars must next year, perhaps, or within a short while, in certain cases, be redredged. It is not a permanent improvement, and under present conditions it never can be a permanent improvement. You are taking out small quantities, it is true, but the headwaters all around, over thousands of square miles, are furnishing to that stream ever-increasing quantities to quickly obliterate the little bit that you take out.

The dredgers at Knoxville and down at Chattanooga, Tenn., take out great quantities of sand and gravel for building purposes. They are perpetually dredging. A firm owning a little section consisting of a few acres dredges thousands of cubic yards of sand and gravel from it, but he never gets to the bottom of it. He never exhausts the supply. The river is perpetually rolling it along its bottom and filling the dredged spot up again, so that it is just as full to-day as it was ten years ago, before he took a yard out, and it will be just as full a hundred years hence. Indeed, it will be fuller a hundred years hence if present conditions continue, although he dredges constantly all the time.

What is the use of spending millions of dollars-and the Government has spent five, six, or seven millions on the Tennessee River already-in making improvements of that sort, many of which are bound to be as short lived as those are? There can be no permanent improvement, no permanent maintenance, of a navigable channel there under present conditions by such methods. It is an unending task; and not only unending, but a task that, as the years go by, becomes a more and more difficult one, because of the annual increase in the quantity of material that is being carried in there. I could give detailed instances covering changes that have gone on in this river, showing that there is a vast quantity of material lodging along the way and accumulating in the river. It is overloaded with the waste from the steep deforested mountain slopes.

Going to the Coosa River, on the other hand, its headwaters in north Georgia are largely forested. It is navigable up to the very foot of the mountains. You can almost throw a stone from the last landing on one of the headwaters of the Coosa River against the steeply rising mountain slope; certainly you could fire a rifle against it. Yet the river, even up there, is navigable, and has been for years. It needs some snags taken out, it is true, and improvements of that sort now and then, but there is not a great accumulation of sand and gravel there, because the mountain slopes up above are all forested land and do not furnish enormous quantities of sand and gravel.

I went down that river 165 miles from Rome, Ga., examining it, and found that there were no accumulations of sand and gravel in it. That was the very thing that I would have expected, because the mountain headwaters there are protected by forests. Just in proportion as the headwaters of that stream are denuded of their forest covering (most of them are entirely too steep to be agricultural lands), you will find sand and gravel accumulating in the stream, and you will find the people of Georgia and Alabama clamoring for larger and larger annual appropriations to clear out the stream and maintain the navigable conditions that they have down there to-day.

Coming around on the eastern side, I could give you illustrations of numerous other streams-the Chattahoochee,, the Savannah, the Broad, the Catawba, the Yadkin, the Roanoke, and the James. They are all affected by the same conditions. There is an enormous quantity of sand and gravel being carried in all of these streams that flow down to the east to the Atlantic. The Yadkin is typical. It is about halfway between the north and the south limits mentioned. The Yadkin channel is rapidly filling with sand. It is accumulating behind any chance obstruction. It has already shallowed the stream and has caused a smaller flood discharge than formerly to spread