at all where the protection is adequate. Of course, even bushes are better than nothing. Bushes may grow up, and are better than the open slope, unprotected, but they are not anything like as good as forests. Moreover, in the open, after the forest has been cut off, fire is more likely to occur. In fact, that is where fire generally occurs. The fires that ravage our forests take place in the open areas which have been cut. They burn over the ground and destroy the soil, and thereby reduce its power to prevent erosion.

Besides that, the sun shines down on the open areas and bakes the ground hard, until it is almost like a tinder box in the summer time, and it is much more likely to take fire and much more likely to be eroded. Then the freezing and thawing (which are much more frequent and much more liable to take place in the open than in the forest) open up the ground, as everybody knows. In the spring, when the frost comes out of the ground, we all know how loose it is and how easy it is to carry it away. That is the same action that takes place on the open slopes, where they are not protected. The freezing and thawing loosen the ground; and in the spring, when the floods come, it is very easy for them to carry it away.

The action of the forest, then, is extremely simple. In the summer time it retards the flowing off of the water. In the winter time it retards the melting of the snow. Of course you may imagine combinations of circumstances, as I have said before, where from a forested area you may get as large a flood as you would have gotten from an unforested area. But I do not think anybody can ever make the intelligent people believe that if you retard the melting of the snow and make it take six months to melt instead of four you are going to have anything but a favorable influence on the floods of our streams or on the amount of sediment which is carried by our streams. With reference to the action and the benefit of forests on the navigation of streams, I maintain that there is a practical unanimity of opinion. I should like to refer briefly to some of the various reports on the subject.

In Mr. Moore's report, on pages 17 to 19, there is a long quotation from Mr. Belgrand and Mr. Valles in regard to the effect of forests in France. Those two men are the two French engineers who have mainly criticised the forests. They wrote their papers in the fifties. Mr. Belgrand wrote his in 1855 and Mr. Valles (whose book I have here) wrote his in 1857. Those two men agree with us regarding the beneficial influence of the forests on the steep slopes which are not suited for cultivation.

Let me read just one sentence from what Mr. Belgrand says, to show you what he does say:

The operation of reforesting is, therefore, excellent when it is practically possible, although it appears demonstrated that the deforesting of the basin of the Seine can not be considered as one of the causes which have contributed to increase or to decrease the height and the number of the floods.

I will refer to that again. That was his idea, which is not shared by many. Most engineers will probably differ from Belgrand in this statement, which was written thirty-five years ago:

But the woods diminish very notably the volume of earthy material transported by the rivers, because they prevent the erosion of land; and it must be recognized that the impoverishment of the earth is much more to be regretted than the disasters caused by inundations.

That is the statement of Mr. Belgrand, who is quoted by those who seem to be in opposition to us, although they really are not.

I have here the book of Mr. Valles, whose paper is quoted by Professor Moore. Here is what he says:

If the demand is limited to the reforesting of the summits and the uncultivated slopes, we wish well of it; but less on account of the diminution of the quantity of water discharged than on account of preventing the erosion of the ground.

If, therefore, we are to understand in this way the advocates of the planThat is to say, as being in favor of the reforesting of the steep slopes which are not used for cultivation

We will not hesitate to join them; but if their claims are more extended, if they wish us to reestablish on a former basis a state of things which has been favorably modified, if they wish to have us prefer the Druidical forests to our modern farms, trees to wheat-we say for the third time, reforesting will not do.

Nobody wants to have the farms sacrificed for woods. We wish to have the forests protected on the mountains and the steep hills which are not suited for cultivation.

One writer who has been much quoted in Mr. Moore's report and in Mr. Chittenden's report is Mr. Lauder, of Vienna. Let me read what Mr. Moore says in that regard:

*

Mr. Ernest Lauder, chief of the Hydrographic Bureau of the Austrian Government, recently made an exhaustive investigation of the records of the Danube, the great river of central Europe. His conclusions are that progressive deforestation of the country has had no effect in increasing the frequency of floods or in augmenting their height. Among other things he showed that the flood of 1899, which was a summer flood, was severest where it came from the heavily wooded districts.

It is just as well to give the whole of a man's idea if you wish to get truly his position in the matter. I have here a more recent paper of Mr. Lauder in connection with the congress in Milan in 1905, and here is what he says:

The forest exerts an influence in any case on the flow of water.

The retention of the water precipitated is, in a certain measure, greater in the more than in the less wooded basin.

For rains, whose importance exceeds certain limits-in times of floods, for example the retention in the more wooded basin becomes more intense than in the one less rich in forest lands

This translation was made abroad. I read it just as it has been translated.

That is to say, that after a certain degree of saturation has been reached the excess of water which the forest held before is set free quite more sensibly, and that after a dry season the influence of rains is felt more rapidly and more progressively in the basin which is less rich in forests, while the reserve takes place in the one where the wooded surface is greater.

It may be stated in closing that the preceding conclusions are entirely in accord with the result of the studies which the Central Hydrographic Office, which is under the author's direction, obtained by examining the question of the influence of the forest on the formation and regimen of high flood waters, and which is opposed to the fact, long considered to be true, that to the presence of forests must be attributed, under all circumstances, a favorable influence on the moderation and even on the prevention of catastrophes due to high water, and to the cutting down of forests a similar influence on the production of such catastrophes or on the quickness of their succession.

A final judgment on the subject of the influence of forests on the regimen of streams can not yet be uttered, the experimental data possessed so far covering only a relative short space of time. Hence, it is important to continue, in the domain of hydrologic science, the researches undertaken up to the present

time, and to complete and investigate thoroughly the ideas given above, the boundaries of which may have to be even closer drawn. Nevertheless, it is believed that this study is useful in this way, that it points out a track to be followed in order to draw near gradually to the end, still so far off, of a knowledge of the movement of the discharges of rivers.

If, now, the final judgment on the subject of the influence of forests on the regimen of streams be unfavorable to the forest to this extent, that there are denied to it certain of the properties attributed to it generally, it does not follow from this that it is necessary to oppose the rewooding of arid surfaces, the replanting of the basins of streams, or the maintenance of plantations of trees. The general utility of the forest is so well settled, the extraordinary appreciation in which it is held as a means of protecting the soil against landslides is so firmly established, its great advantageousness

As they express it

especially for the spring district, in holding back earth thrusts and reducing the amount of sediment carried by rivers so important, that these reasons alone justify fully the great possible promotion of forest culture.

That is what Mr. Lauder said the authority who is quoted by those who apparently are opposed to us in this matter.

Mr. LEVER. And this is a more recent paper?

Mr. SWAIN. This is a more recent paper.

Mr. Cippolletti, who summed up the conclusions of all the writers who contributed to that congress, say this:

But with regard to the régime of high-water flow and ordinary floods there is no material difference of opinion. In such localities the waters running off the surface unite with the springs in providing the supply of water for a river. Thus, all the writers agree that forests exert a moderating influence on the run-off of surface waters, owing to a large proportion of the water being retained by the leaves and other parts of the plants, also on account of a quantity of it being absorbed by the layers of dead leaves, moss, and humus which form the top covering of the forest ground, and partly also to the obstruction which roots above ground form to the rapid flow-off of the surface water by forcing it to remain stagnant in a thin sheet, instead of accumulating in a mass and running off quickly in the shape of brooks, which is what happens on lands where the surface water finds little obstruction and is apt to produce erosion. To this may be added, in the case of cold climates, the additional advantage that the snow lies longer in the forest and melts here more slowly than in the open country. To conclude, forests act as real regulators, obliging the rain water to flow much more slowly to the bottom of the valley than it would do otherwise, and by this means insuring a more uniform and continuous flow in the lower reaches of rivers.

Can the destruction of forests bring about a great deterioration and even the total loss of the layer of cultivable soil of cultivable lands, of those which are next to them or lying beneath them, and, even further off, of the level portions of large valleys? Can it also cause landslips, landslides, and avalanches?

Upon this point it will be very much more easy for me to arrive at a conclusion, because the authors of the papers, without distinction, and all technical experts generally admit that the deforestation of sloping lands, especially if it is followed by a breaking up and cultivation of the soil, will cause the damages and injuries enumerated at the head of this chapter.

I have other authorities that I can cite.

So that I think it is fair to say-and I am fairly familiar with the literature of this subject-that the authorities throughout the world are generally agreed on the importance of the forests on the steep slopes.

Professor Moore's fifth conclusion is that forests may cause excessive precipitation. Of course the ultimate source of all floods is the rainfall; but it does not follow from that that there is nothing which modifies that precipitation and affects the flow of the streams.

Then he says that compared with the total area of a given watershed, that of the headwaters is usually small, and therefore it will

not affect the flow. There is a confusion in many minds with reference to that; and Mr. Chittenden falls into the same confusion of reasoning. Mr. Chittenden argues that the rivers do not overflow up in the mountains; that they overflow down in the lowlands. Of course we all know that. It is not that we want to prevent the overflow of the rivers in the mountains; but it is the quick discharge of the streams from the mountain slopes that makes the floods in the lowlands. The water rushes down these mountain slopes and collects in the reaches of the rivers below which the slope is so gradual that it can not carry it off. Therefore, the floods from these high mountain districts, even though they might be comparatively small, cause the floods in the rivers-in the Ohio River at Pittsburg and Cincinnati, for instance, just as much as at Louisville and Memphis and New Orleans, where the floods are dangerous. Pittsburg is away up near the headwaters in comparison with the other places.

The CHAIRMAN. Pardon me for one moment while you are on that point, Mr. Swain, because it is a very important one. If we should concede that in the case of the forested area at the headwaters of the stream, no water would come into the river as a result of rain, whereas if the woods were cut away all the water that fell would immediately come in, your conclusion, of course, would be inevitable. But, of course, such a violent presumption as that is not in your mind?

Mr. SWAIN. Not at all.

The CHAIRMAN. You will concede that part of the rain that falls even on a forested area will come into the streams?

Mr. SWAIN. Certainly.

The CHAIRMAN. And the question is whether the comparatively slight difference between what would come into the streams if the forest were there and what would come into them if the forest were not there would result in floods in the lower reaches of the river.

Mr. SWAIN. Our point is in reference to the suddenness of the discharge. We say that it is not a question of the slight difference of the total amount which would come into the streams. We believe that on the whole, taking the springs into account, there is more that comes in from the forested areas than from the deforested areas. But that is a minor matter. The question is the suddenness of the water flowing in. Of course, if you had a very sudden shower on a rocky surface inclined like a house roof, it would all go in, whereas in the forest some of it would be held back and evaporated. That is true. There would more go in that case; but in the forest it is held back. It is not discharged suddenly into the streams, and therefore, on the whole, the floods are diminished.

But, more than that, more important than whether or not the floods are diminished, is the question of the erosion which carries the material down. We all know how even a small rainfall-a summer rain which sometimes in the mountains comes down pretty rapidly, in sheets; a rain which would not of itself cause a flood in the streams at all-will on a deforested area, on open ground, carry away a great deal of earth with it. These summer rains come down just as severely as any other rains. In fact, they come down more suddenly. Very likely there is more rain discharged in a given time in the summer than at any other time.

The local showers, the thunder showers which come down only over a small area, and which would not cause a flood in the streams at all, carry the earth with them just as much as the longer rains which cause the floods.

Just one point more: Mr. Moore's seventh conclusion is that "the run-off of our rivers is not materially affected by any other factor than the precipitation." In other words, the slope, the character of the ground, the character of the rock, whether there are forests there or not, make no difference in the run-off of our streams; it is simply a question of the precipitation. That seems to me a perfectly absurd statement to make, unless it is qualified.

Then his eighth and ninth conclusions are:

The high waters are not higher and the low waters are not lower than formerly.

Floods are not of greater frequency and longer duration than formerly. Mr. Moore has previously said that we have not data enough to draw definite conclusions. He says:

All of these problems could be definitely settled beyond the possibility of argument if we had accurate river gaugings from day to day and year to year, together with a full knowledge of the rainfall and of the proportion of the wooded to cleared areas data that unfortunately we do not have. We must, therefore, reason empirically from the best information at hand; and this insufficiency of data renders less positive the conclusions of all investigators, no matter which side of the question they may be on.

Mr. LEVER. When did he say that?
Mr. SWAIN. In this paper.

Mr. LEVER. This same paper here?

Mr. SWAIN. Yes. And yet he positively draws the conclusion that the high waters are not higher and the low waters are not lower. He ought logically to draw the conclusion that he can not tell whether they are higher or lower. And he does not touch the question of erosion, which is the question on which our case rests.

In regard to the Seine, I want to say just one word. We have recently had a great object lesson in Paris. I have an interesting quotation here which I should like to read to you. It is from the Emperor Julian, and is an extract from one of his works written in the fourth century. Emperor Julian was for six years stationed in Gaul and his headquarters were in Paris. He writes as follows:

I was formerly in winter quarters at my dear Lutetia, for that is what the Gauls call the town of the Parisii. It is an island of no great size, situated in the river, which surrounds it on every side. Bridges built on piles lead to it on both sides. The river seldom falls or rises, but is generally the same winter and summer.

Of course, I do not present that as a scientific argument, but it simply shows what he said at that time. He was for six years in Gaul, and he says that the river seldom falls or rises, but is the same summer and winter.

From a paper which I saw last week I cut this out:

M. Velain, professor of physical geography at the Sorbonne, has drawn up a report on the causes of the French floods, which he attributes, in the first place, to exceptional rainfall during the last six months, and, secondly, to the wholesale destruction of forests. In this respect Mr. Velain voices the unanimous

opinion of French scientists.