by help of which he can collect together in a few moments all the texts upon any particular subject, such as the sea, the wind, the sky, human life, the shadows of evening. The study of the Bible is not one which I should recommend to very young Japanese students, because of the quaintness of the English. Before a good knowledge of English forms is obtained, the archaisms are apt to affect the students' mode of expression. But for the advanced student of literature, I should say that some knowledge of the finest books in the Bible is simply indispensable. The important books to read are not many. But one should read at least the books of Genesis, Exodus, Ruth, Esther, the Song of Songs, Proverbs and, above all, Job. Job is certainly the grandest book in the Bible; but all of those which I have named are books that have inspired poets and writers in all departments of English literature to such an extent that you can scarcely read a masterpiece in which there is not some conscious or unconscious reference to them. Another book of philosophical importance is Ecclesiastes, where, in addition to much proverbial wisdom, you will find some admirable world-poetry-that is, poetry which contains universal truth about human life in all times and all ages. Of the historical books and the law books I do not think that it is important to read much; the literary element in these is not so pronounced. It is otherwise with the prophetic books, but here in order to obtain a few jewels of expression, you have to read a great deal that is of little value. Of the New Testament there is very little equal to the old in literary value; indeed, I should recommend the reading only of the closing book-the book called the Revelation, or the Apocalypse, from which we have derived a literary adjective | "apocalyptic," to describe something at once very terrible and very grand. Whether one understands the meaning of this mysterious text makes very little difference; the sonority and the beauty of its sentences, together with the tremendous character of its imagery, can not but powerfully influence mind and ear, and thus stimulate literary taste. At least two of the great prose writers of the nineteenth century, Carlyle and Ruskin, have been vividly influenced by the book of the Revelation. Every period of English literature shows some influence of Bible study, even from the old Anglo-Saxon days; and during the present year, the study has so little slackened that one constantly sees announcements of new works upon the literary elements of the Bible. Perhaps one of the best is Professor Moulton's "Modern Reader's Bible," in which the literary side of the subject receives better consideration than in any other work of the kind published for general use. If this brief lecture has shown the real place of the King James version in English literature, and suggested to you the reason why the book has an all-important value, independently of any religious thought in it-quite sufficient has been said. It would be of no use whatever to spend the time otherwise utilizable, in pointing out beauties of the text. What beauty there is is of a kind so simple that explanation is quite unnecessary. Where I think that the value of the reading would be greatest for you, is in regard to measure and symmetry and euphony in English construction. But that means a great deal so much that the best illustration of it is the observation already made, that all English written since the sixteenth century has been colored by the Bible. THE RELATION OF FORESTS TO STREAM CONTROL1 GIFFORD PINCHOT As Gifford Pinchot (1865-) was the first man to do systematic work in forestry in the United States. For twelve years he was connected with the national government as chief of the Bureau of Forestry, and labored diligently for the conservation of our national resources. a result of political differences with his superior, Secretary Ballinger of the Interior, he was dismissed from office in 1910, and soon became instrumental in the formation of the Progressive Party. Since 1903 he has been a professor at Yale, and since 1910, the president of the National Conservation Association. The following article, published in 1908, is an example of straightforward scientific exposition. courses of many of our rivers, but it cannot and does not claim to regulate in the least the water supply of the streams. The method of storage reservoirs, extensively tried in France, has been suggested as a method of river improvement in the United States. Reservoirs filled in the spring freshet season serve to increase the flow later in the year when the streams run low. Floods may thus be prevented, and the immense loads of silt which they would otherwise have brought down are thus kept from being dropped by the slow current in the lower channel. Theoretically this method of storage reservoirs will accomplish all that can be desired in regulating stream flow and preventing excessive deposition, if only adequate storage capacity is available. In practice, too, it will doubtless be efficient in places where the erosion is not rapid. But the great disadvantage of this method, as is proved by the ex THE phenomenal development of industry and the consequent increased demand for transportation have turned attention to our most natural means of inland transportation-the lakes and rivers. It has forced us to realize that our streams, in spite of the tens of millions of dollars appropriated for their development, are becoming less navigable. Increasing amounts of sediment are deposited each year in their middle and lower courses, while the flow of the streams themselves becomes less regular. Navigable with difficulty, if at all, during the summer, they become turbulent and turbid during the spring, overflow their banks, and often carry destruction to life and property. The skill of our engineers is taxed to the utmost to keep harbors and rivers free from the constantly recurring deposits of sediment. Because of the rapidly increasing tonnage and draft of vessels, it is not sufficient merely to maintain the present depth of our riv-perience of the French engineers, lies in ers and harbors. Their depth must be constantly increased or they will gradually fail to accommodate the larger vessels, and such of them as fail must finally be abandoned altogether. More powerful dredging machinery is constantly coming into use. Efforts are common to prevent the deposit of sediment by confining streams to channels narrow enough to accelerate the current and so lessen the rate of deposition. This method of channel adjustment has accomplished great good in improving the Reprinted by permission from The Annals of the American Academy of Political and Social Science, January, 1908, Vol. XXXI, the fact that the reservoirs themselves become clogged with detritus and must sooner or later, varying with the forest conditions and the character of the topography drained, be either abandoned or maintained by constant clearing out at large expense. The engineers of the United States. Reclamation Service fully realize that the amount of solid matter carried by a stream is a very serious problem in connection with the construction of storage reservoirs for irrigation purposes. Streams from barren watersheds abound in violent freshets which carry with them eroded sediment, to be deposited in the first pool of still water they encounter, and thus reduce the storage capacity of the reservoirs into which they flow. Mill dams completely filled with sediment are to be seen everywhere, and offer good demonstrations of the damage to storage reservoirs from silting. The regulation of streams by storage reservoirs is really an imitation of what nature is able to accomplish by the forests. Forests at the sources of the streams are veritable storage reservoirs, and without them no artificial remedy can be either adequate or permanent. Erosion destroys reservoirs, and must be controlled if reservoirs are to succeed. This can be done only by conserving or restoring the forests. The forest cover alone can reduce the amount of sediment carried by water, and make possible the permanent improvement of inland waterways. To check erosion by reforestation, work must begin in the highlands, because there the slopes are steepest, the rainfall greatest, and the action of frost most considerable, and therefore the process of erosion is most rapid and the results most destructive. No one will deny the necessity for engineering methods to cope with the moderate deposits of silt and the seasonal irregularities in flow, which may indeed be lessened by forest cover, but which are unavoidable so long as the sun shines and the rain falls. Yet it remains true that a forest cover interposed between rain and rock affords the best natural means for regulating streams and reducing the loads of detritus. Without such a forest cover every attempt to improve the regimen and the channel of a stream will be little more than a temporary expedient. Both wide experience and scientific investigation have shown that there are two functions exercised by the forest in relation to stream-flow. I. Its tendency to reduce the difference between high and low water, an influence which is of most importance in the distribution of flood crests, and in maintaining a steady flow of water during the different seasons of the year and during cycles of dry and wet years. 2. Its value as a surface protection against soil erosion, thus reducing the solid burden of storm waters, and decreasing the deposits of sand and silt, which are the causes of shallow and changing channels. These two functions follow from the very nature of the forest as a soil cover. The roots of trees penetrate through the soil to the underlying rock, where they fix themselves in the crevices, in this way hold in place the loose soil and prevent slipping and washing. The crowns of the trees break the force of the rain and also protect the soil from being carried away to the lower valleys during heavy storms. The leaves and the branches allow the rain to reach the ground but gradually; after a rain, water continues to drip from the crown for several hours, and the soil is thus enabled to absorb the greater part of it. Screened from the rays of the sun and covered with a surface mulch of fallen leaves and humus, the soil remains loose and granular in structure and is therefore capable of imbibing and retaining water with sponge-like capacity. It is strewn with fallen leaves, branches, and trunks, and traversed by a net-work of dead and live roots which impede the superficial runoff of water after heavy storm. This retardation of the superficial run-off allows more of it to sink into the ground through the many channels left in the soil by decayed roots. Surface run-off of rain water is wasteful and destructive, and unless artificially controlled serves as a rule no useful purpose and may inflict great loss. Sub-surface drainage makes the best use of the total precipitation that reaches the ground. It serves both for the sustenance of plant life and for the flow of streams. Accordingly the agency of the forest cover in increasing the seepage run-off at the expense of the surface run-off is the most important function which the forest performs in relation to water supply. A common conception of the effect of forest destruction upon climate is that it reduces the amount of rainfall. Because springs become dry and streams shrink in a deforested region, it is assumed that less rain must fall. Whether or not there be any truth in this assumption (I believe there is), it is certain that the main cause of the observed facts is the profound effect which forest destruction has upon the course which the water takes after it reaches the ground. The greatest influence of the forest is not upon the amount of rain which falls, but on what becomes of the rain after it falls. The water that sinks into the ground passes for greatly varying distances beneath the surface before reappearing, and is thus drawn off gradually from the forested watershed and supplies the brooks with pure water relatively free from detritus. How active a part is played by the forest in regulating the run-off is clearly shown by actual measurements of the flow of streams which drain forested and unforested watersheds. A typical illustration of streams from barren, treeless watersheds may be found in the flow of Queen Creek, in Arizona. This stream discharges only in violent freshets, recurring usually as great floodwaves which subside almost as soon as they arise. The area of the drainage basin is 143 square miles, of which 61 per cent. is above an elevation of 3,000 feet. The rainfall is estimated to be about 15 inches. The maximum flood discharge of Queen Creek in 1896 was 9,000 cubic feet per second, and the mean discharge was 15 cubic feet per second; during a large portion of the year the stream was entirely dry. Cedar Creek, in Washington, is typical of streams flowing from timbered watersheds. The basin of Cedar Creek lies on the western slope of the Cascade Mountains and is covered with a dense forest and a very heavy undergrowth of ferns and moss. The drainage area is the same as that of Queen Creek, 143 square miles. The precipitation for the year 1897 was about 93 inches for the lower portion of the basin, and probably 150 inches on the mountain summits; in spite, however, of the fact that the precipitation in Cedar Creek basin was from six to nine times more than that in Queen Creek basin, the maximum flood discharge of Cedar Creek for 1897 was but 3,601 cubic feet per second, as against the 9,000 cubic feet of Queen Creek. On the other hand the flow of Cedar Creek was continuous throughout the year, and the minimum discharge was never less than 27 per cent. of the mean of the year. The mean discharge of Cedar Creek was 1,089 cubic feet as against 15 feet for Queen Creek. This radical difference in the behavior of the two streams can be explained only by the difference in the soil cover of the two basins. Cedar Creek basin is covered with a heavy forest, while Queen Creek basin is almost entirely bare, with but a few scattering pinion trees and a little brush or grass. Mr. Marsden Manson, in discussing the stream flow from certain points on the Yuba River basin, California, makes a very interesting comparison between its two branches, North Fork and South Fork, of which the first has a forested and the second a denuded basin. Both of the catchment areas lie on the western slope of the Sierra Nevada, and have exposures of marked similarity. The south branch of the North Fork has a watershed area of 139 square miles, which gave in 1900 a maximum run-off of 113 cubic feet per second, or 0.8 cubic feet per second per square mile. This drainage area is well covered with timber and brush, and for four months gives a minimum run-off of 1,441,125,000 cubic feet. On the South Fork, above Lake Spaulding, there is a watershed of 120 square miles from which the scattering timber that once existed has been cut off. The run-off of this area is practically nothing for four months in each year, because of this absence of forests. If this area was afforested, and gave a minimum run-off of 0.8 cubic foot per second per square mile, the discharge would be 100 cubic feet per second, or equivalent to 1,036,800,00 cubic feet of effective stor age capacity. To supply water for mining and power purposes a number of costly storage reservoirs have been built on the South Fork. By reforesting the small watershed a natural reservoir would be created whose storage capacity would be almost equal to the storage capacity of all the reservoirs above Lake Spaulding dam. A careful study of the behavior of the streamflow on several small timbered and non-timbered catchment areas in the San Bernardino Mountains of Southern California, made by Professor Toumey for the Forest Service in 1902, brought out in a most convincing manner the effect of the forest in decreasing surface run-off and sustaining the flow of mountain streams. Three timbered drainage areas were studied. These gave during December-a month of unusually heavy precipitation a run-off of but 5 per cent. of the heavy rainfall for that month; during the following months of January, February and March, they gave a run-off of approximately 37 per cent. of the total precipitation, and three months after the close of the rainy season still supported a well-sustained streamflow. At the same time, the similar and neighboring nontimbered catchment area under observation gave during December a run-off of 40 per cent. of the rainfall, and during the three following months a run-off of 95 per cent. In April the run-off was less than one-third of that from each of the forest catchment areas, and in June the stream from the non-forested area was dry. Streams flowing from barren, treeless watersheds carry an amount of gravel, sand and soil which is simply enormous compared to the amount in streams from timbered areas. Thus the United States Geological Survey determined the amount of silt carried by the Gila River at the Buttes, a stream whose basin and regimen is similar to that of Queen Creek, of Arizona, to be 10 per cent. of the volume wet or 2 per cent. of solids. To appreciate these figures it must be remembered that one-fourth of one per cent. of solid burden in the stream is enough to make the water turbid. As long as the ground is protected by a natural covering of forest growth, rainfall has very little erosive action. It is only after the ground is laid bare by the removal of the forest that the erosion of the soil attains dangerous proportions. There has, of course, always been, even when the natural forests were unimpaired, some erosion, especially in the watersheds. of streams in the Southeast and Southwest, but not to the extent which now obtains, and the present erosion is not only excessive, but is yearly increasing. It is the price, and in a large measure the product, of necessary agricultural and industrial development under defective methods of work. According to studies. of Humphreys and Abbott the wearing down of the earth's surface over a region such as the Mississippi Valley is something like one foot in five thousand years, independent of human action. At such a rate of erosion the amount of sediment carried by the Mississippi River before the dawn of civilization could not be more than 70,000,000 tons per year. According to Professor Shaler the wearing down of the Mississippi Valley under complete tillage will be about the same as that of the Valley of the Po in northern Italy, or one foot in one thousand years. such a rate of erosion, the solid burden of the Mississippi River should be 350,000,000 tons. But the amount of solid matter carried every year by the Mississippi River was estimated several years ago to be 400,000,000 tons. In other words, the erosion had then reached, if not exceeded, that of the Po Valley. It is greater now. The formation of soil through underground decay of the rocks cannot keep pace with such a rate of erosion. Unless measures are taken to check it the fertile layer of soil must gradually disappear, as has happened already over large areas in the Old World from precisely similar causes. At The ruinous effects of the destruction of mountain forests upon the navigability of streams and the cultural results of |