Chapter III · (continued 2)

The first questions which suggest themselves upon this statement are: what becomes of this immense quantity of water and from what source does the tree derive it We are told in reply that it is absorbed from the air by the humus and mineral soil of the wood, and supplied again to the tree through its roots, by a circulation analogous to that observed in Ward's air-tight cases. When we recall the effect produced on the soil even of a thick wood by a rain-fall of one inch, we find it hard to believe that two hundred and forty times that quantity, received by the ground between early spring and autumn, would not keep it in a state of perpetual saturation, and speedily convert the forest into a bog.

No such power of absorption of moisture by the earth from the atmosphere, or anything approaching it, has ever been shown by experiment, and all scientific observation contradicts the supposition. Schubler found that in seventy-two hours thoroughly dried humus, which is capable of taking up twice its own weight of water in the liquid state, absorbed from the atmosphere only twelve per cent. of its weight of humidity; garden-earth five and one-fifth per cent. and ordinary cultivated soil two and one-third per cent. After seventy-two hours, and, in most of his experiments with thirteen different earths, after forty-eight hours, no further absorption took place. Wilhelm, experimenting with air-dried field-earth, exposed to air in contact with water and protected by a bell-glass, found that the absorption amounted in seventy-two hours to two per cent. and a very small fraction, nearly the whole of which was taken up in the first forty-eight hours. In other experiments with carefully heat-dried field-soil, the absorption was five per cent. in eighty-four hours, and when the water was first warmed to secure the complete saturation of the air, air-dried garden-earth absorbed five and one-tenth per cent. in seventy-two hours.

In nature, the conditions are never so favorable to the absorption of vapor as in those experiments. The ground is more compact and of course offers less surface to the air, and, especially in the wood, it is already in a state approaching saturation. Hence, both these physicists conclude that the quantity of aqueous vapor absorbed by the earth from the air is so inconsiderable "that we can ascribe to it no important influence on vegetation." [Footnote: Wilhelm, Der Boden und das Wasser, pp. 14,20.] Besides this, trees often grow luxuriantly on narrow ridges, on steep declivities, on partially decayed stumps many feet above the ground, on walls of high buildings, and on rocks, in situations where the earth within reach of their roots could not possibly contain the tenth part of the water which, according to Schleiden and Pfaff, they evaporate in a day. There are, too, forests of great extent on high bluffs and well-drained table-lands, where there can exist, neither in the subsoil nor in infiltration from neighboring regions, an adequate source of supply for such consumption. It must be remembered, also, that in the wood the leaves of the trees shade each other, and only the highest stratum of foliage receives the full influence of heat and light; and besides, the air in the forest is almost stagnant, while in the experiments of Unger, Marshal, Vaillant, Pfaff and others, the branches were freely exposed to light, sun, and atmospheric currents. Such observations can authorize no conclusions respecting the quantitative action of leaves of forest trees in normal conditions.

Further, allowing two hundred days for the period of forest vital action, the wood must, according to Schleiden's position, exhale a quantity of moisture equal to an inch and one-fifth of precipitation per day, and it is hardly conceivable that so large a volume of aqueous vapor, in addition to the supply from other sources, could be diffused through the ambient atmosphere without manifesting its presence by ordinary hygrometrical tests much more energetically than it has been proved to do, and in fact, the observations recorded by Ebermayer show that though the RELATIVE humidity of the atmosphere is considerably greater in the cooler temperature of the wood, its ABSOLUTE humidity does not sensibly differ from that of the air in open ground. [Footnote: Ebermeyer, Die Physikalischen, Einwirkungen des Waldes, i., pp. 150 et seqq. It may be well here to guard my readers against the common error which supposes that a humid condition of the AIR is necessarily indicated by the presence of fog or visible vapor. The air is rendered humid by containing INVISIBLE vapor, and it becomes drier by the condensation of such vapor into fog, composed of solid globules or of hollow vesicles of water—for it is a disputed point whether the particles of fog are solid or vesicular. Hence, though the ambient atmosphere may hold in suspension, in the form of fog, water enough to obscure its transparency, and to produce the sensation of moisture on the skin, the air, in which the finely divided water floats, may be charged with even less than an average proportion of humidity.]

The daily discharge of a quantity of aqueous vapor corresponding to a rain-fall of one inch and a fifth into the cool air of the forest would produce a perpetual shower, or at least drizzle, unless, indeed, we suppose a rapidity of absorption and condensation by the ground, and of transmission through the soil to the roots and through them and the vessels of the tree to the leaves, much greater than has been shown by direct observation. Notwithstanding the high authority of Schleiden, therefore, it seems impossible to reconcile his estimates with facts commonly observed and well established by competent investigators. Hence the important question of the supply, demand, and expenditure of water by forest vegetation must remain undecided, until it can be determined by something approaching to satisfactory direct experiment. [Footnote: According to Cezanne, Surrell, Etude sur les Torrents, 2e edition, ii., p. 100, experiments reported in the Revue des Eaux et Forets for August, 1868, showed the evaporation from a living tree to be "almost insignificant." Details are not given.]

Balance of Conflicting Influences of Forest on Atmospheric Heat and Humidity.

We have shown that the forest, considered as dead matter, tends to diminish the moisture of the air, by preventing the sun's rays from reaching the ground and evaporating the water that falls upon the surface, and also by spreading over the earth a spongy mantle which sucks up and retains the humidity it receives from the atmosphere, while, at the same time, this covering acts in the contrary direction by accumulating, in a reservoir not wholly inaccessible to vaporizing influences, the water of precipitation which might otherwise suddenly sink deep into the bowels of the earth, or flow by superficial channels to other climatic regions. We now see that, as a living organism, it tends, on the one hand, to diminish the humidity of the air by sometimes absorbing moisture from it, and, on the other, to increase that humidity by pouring out into the atmosphere, in a vaporous form, the water it draws up through its roots. This last operation, at the same time, lowers the temperature of the air in contact with or proximity to the wood, by the same law as in other cases of the conversion of water into vapor.

As I have repeatedly said, we cannot measure the value of any one of those elements of climatic disturbance, raising or lowering of temperature, increase or diminution of humidity, nor can we say that in any one season, any one year, or any one fixed cycle, however long or short, they balance and compensate each other. They are sometimes, but certainly not always, contemporaneous in their action, whether their tendency is in the same or in opposite directions, and, therefore, their influence is sometimes cumulative, sometimes conflicting; but, upon the whole, their general effect is to mitigate extremes of atmospheric heat and cold, moisture and drought. They serve as equalizers of temperature and humidity, and it is highly probable that, in analogy with most other works and workings of nature, they, at certain or uncertain periods, restore the equilibrium which, whether as lifeless masses or as living organisms, they may have temporarily disturbed. [Footnote: There is one fact which I have nowhere seen noticed, but which seems to me to have an important bearing on the question whether forests tend to maintain an equilibrium between the various causes of hygroscopic action, and consequently to keep the air within their precincts in an approximately constant condition, so far as this meteorological element is concerned. I refer to the absence of fog or visible vapor in thick woods in full leaf, even when the air of the neighboring open grounds is so heavily charged with condensed vapor as completely to obscure the sun. The temperature of the atmosphere in the forest is not subject to so sudden and extreme variations as that of cleared ground, but at the same time it is far from constant, and so large a supply of vapor as is poured out by the foliage of the trees could not fail to be sometimes condensed into fog by the same causes as in the case of the adjacent meadows, which are often covered with a dense mist while the forest-air remains clear, were there not some potent counteracting influence always in action. This influence, I believe, is to be found partly in the equalization of the temperature of the forest, and partly in the balance between the humidity exhaled by the trees and that absorbed and condensed invisibly by the earth.] When, therefore, man destroys these natural harmonizors of climatic discords, he sacrifices an important conservative power, though it is far from certain that he has thereby affected the mean, however much he may have exaggerated the extremes of atmospheric temperature and humidity, or, in other words, may have increased the range and lengthened the scale of thermometric and hygrometric variation.

Special Influence of Woods on Precipitation.

With the question of the action of forests upon temperature and upon atmospheric humidity is intimately connected that of their influence upon precipitation, which they may affect by increasing or diminishing the warmth of the air and by absorbing or exhaling uncombincd gas and aqueous vapor. The forest being a natural arrangement, the presumption is that it exercises a conservative action, or at least a compensating one, and consequently that its destruction must tend to produce pluviometrical disturbances as well as thermometrical variations. And this is the opinion of perhaps the greatest number of observers. Indeed, it is almost impossible to suppose that, under certain conditions of time and place, the quantity and the periods of rain should not depend, more or less, upon the presence or absence of forests; and without insisting that the removal of the forest has diminished the sum-total of snow and rain, we may well admit that it has lessened the quantity which annually falls within particular limits. Various theoretical considerations make this probable, the most obvious argument, perhaps, being that drawn from the generally admitted fact, that the summer and even the mean temperature of the forest is below that of the open country in the same latitude. If the air in a wood is cooler than that around it, it must reduce the temperature of the atmospheric stratum immediately above it, and, of course, whenever a saturated current sweeps over it, it must produce precipitation which would fall upon it, or at a greater or less distance from it.

We must here take into the account a very important consideration. It is not universally or even generally true, that the atmosphere returns its condensed humidity to the local source from which it receives it. The air is constantly in motion,

—howling tempests scour amain From sea to land, from land to sea;

[Footnote: Und Sturme brausen um die Wette Vom Meer aufs Land, vom Land aufs Meer. Goethe, Faust, Song of the Archangels.]

and, therefore, it is always probable that the evaporation drawn up by the atmosphere from a given river, or sea, or forest, or meadow, will be discharged by precipitation, not at or near the point where it rose, but at a distance of miles, leagues, or even degrees. The currents of the upper air are invisible, and they leave behind them no landmark to record their track. We know not whence they come, or whither they go. We have a certain rapidly increasing acquaintance with the laws of general atmospheric motion, but of the origin and limits, the beginning and end of that motion, as it manifests itself at any particular time and place, we know nothing. We cannot say where or when the vapor, exhaled to-day from the lake on which we float, will be condensed and fall; whether it will waste itself on a barren desert, refresh upland pastures, descend in snow on Alpine heights, or contribute to swell a distant torrent which shall lay waste square miles of fertile corn-land; nor do we know whether the rain which feeds our brooklets is due to the transpiration from a neighboring forest, or to the evaporation from a far-off sea. If, therefore, it were proved that the annual quantity of rain and dew is now as great on the plains of Castile, for example, as it was when they were covered with the native forest, it would by no means follow that those woods did not augment the amount of precipitation elsewhere. The whole problem of the pluviometrical influence of the forest, general or local, is so exceedingly complex and difficult that it cannot, with our present means of knowledge, be decided upon a priori grounds. It must now be regarded as a question of fact which would probably admit of scientific explanation if it were once established what the actual fact is.

Unfortunately, the evidence is conflicting in tendency, and sometimes equivocal in interpretation, but I believe that a majority of the foresters and physicists who have studied the question are of opinion that in many, if not in all cases, the destruction of the woods has been followed by a diminution in the annual quantity of rain and dew. Indeed, it has long been a popularly settled belief that vegetation and the condensation and fall of atmospheric moisture are reciprocally necessary to each other, and even the poets sing of

Afric's barren sand, Where nought can grow, because it raineth not, And where no rain can fall to bless the land, Because nought grows there.

[Footnote: Det golde Strog i Afrika, Der Intet voxe kan, da ei det regner, Og, omvendt, ingen Regn kan falde, da Der Intet voxer. Paudan-Muller, Adam Hamo, ii., 408.]

Before going further with the discussion, however, it is well to remark that the comparative rarity or frequency of inundations in earlier or later centuries is not necessarily, in most cases not probably, entitled to any weight whatever, as a proof that more or less rain fell formerly than now; because the accumulation of water in the channel of a river depends far less upon the quantity of precipitation in its valley, than upon the rapidity with which it is conducted, on or under the surface of the ground, to the central artery that drains the basin. But this point will be more fully discussed in a subsequent chapter.

In writers on the subject we are discussing, we find many positive assertions about the diminution of rain in countries which have been stripped of wood within the historic period, but these assertions very rarely rest upon any other proof than the doubtful recollection of unscientific observers, and I am unable to refer to a single instance where the records of the rain-gauge, for a considerable period before and after the felling or planting of extensive woods, can be appealed to in support of either side of the question. The scientific reputation of many writers who have maintained that precipitation has been diminished in particular localities by the destruction of forests, or augmented by planting them, has led the public to suppose that their assertions rested on sufficient proof. We cannot affirm that in none of these cases did such proof exist, but I am not aware that it has ever been produced. [Footnote: Among recent writers, Clave, Schacht, Sir John F. W. Herschel, Hohenstein, Barth, Asbjornsen, Boussingault, and others, maintain that forests tend to produce rain and clearings to diminish it, and they refer to numerous facts of observation in support of this doctrine; but in none of these does it appear that these observations are supported by actual pluviometrical measure. So far as I know, the earliest expression of the opinion that forests promote precipitation is that attributed to Christopher Columbus, in the Historie del S. D. Fernando Colombo, Venetia, 157l, cap. lviii., where it is said that the Admiral ascribed the daily showers which fell in the West Indies about vespers to "the great forests and trees of those countries," and remarked that the same effect was formerly produced by the same cause in the Canary and Madeira Islands and in the Azores, but that "now that the many woods and trees that covered them have been felled, there are not produced so many clouds and rains as before."

Mr. H. Harrisse, in his very learned and able critical essay, Fernand Colomb, sa Vie et ses Oeuvres, Paris, 1872, has made it at least extremely probable that the Historie is a spurious work. The compiler may have found this observation in some of the writings of Columbus now lost, but however that may be, the fact, which Humboldt mentions in Cosmos with much interest, still remains, that the doctrine in question was held, if not by the great discoverer himself, at least by one of his pretended biographers, as early as the year 1571.]

The effect of the forest on precipitation, then, is by no means free from doubt, and we cannot positively affirm that the total annual quantity of rain is even locally diminished or increased by the destruction of the woods, though both theoretical considerations and the balance of testimony strongly favor the opinion that more rain falls in wooded than in open countries. One important conclusion, at least, upon the meteorological influence of forests is certain and undisputed: the proposition, namely, that, within their own limits, and near their own borders, they maintain a more uniform degree of humidity in the atmosphere than is observed in cleared grounds. Scarcely less can it be questioned that they tend to promote the frequency of showers, and, if they do not augment the amount of precipitation, they probably equalize its distribution through the different seasons. [Footnote: The strongest direct evidence which I am able to refer to in support of the proposition that the woods produce even a local augmentation of precipitation is furnished by the observations of Mathieu, sub-director of the Forest-School at Nancy. His pluviometrical measurements, continued for three years, 1866-1868, show that during that period the annual mean of rain-fall in the centre of the wooded district of Cinq-Tranchees, at Belle Fontaine on the borders of the forest, and at Amance, in an open cultivated territory in the same vicinity, was respectively as the numbers 1,000, 957, and 853.

The alleged augmentation of rain-fall in Lower Egypt, in consequence of large plantations by Mehemet Ali, is very frequently appealed to as a proof of this influence of the forest, and this case has become a regular common-place in all discussions of the question. It is, however, open to the same objection as the alleged instances of the diminution of precipitation in consequence of the felling of the forest.

This supposed increase in the frequency and quantity of rain in Lower Egypt is, I think, an error, or at least not an established fact. I have heard it disputed on the spot by intelligent Franks, whose residence in that country began before the plantations of Mehemet Ali and Ibrahim Pacha, and I have been assured by them that meterological observations, made at Alexandria about the begiuning of this century, show an annual fall of rain as great as is usual at this day. The mere fact that it did not rain during the French occupation is not conclusive. Having experienced a gentle shower of nearly twenty-four hours' duration in Upper Egypt, I inquired of the local governor in relation to the frequency of this phenomenon, and was told by him that not of drop of rain had fallen at that point for more than two years previous.

The belief in the increase of rain in Egypt rests almost entirely on the observations of Marshal Marmont, and the evidence collected by him in 1836. His conclusions have been disputed, if not confuted, by Joinard and others, and are probably erroneous. See Foissac, Meteorologie, German translation, pp. 634-639.

It certainly sometimes rains briskly at Cairo, but evaporation is exceedingly rapid in Egypt—as any one who ever saw a Fellah woman wash a napkin in the Nile, and dry it by shaking it a few moments in the air, can testify; and a heap of grain, wet a few inches below the surface, would probably dry again without injury. At any rate, the Egyptian Government often has vast quantities of wheat stored at Boulak in uncovered yards through the winter, though it must be admitted that the slovenliness and want of foresight in Oriental life, public and private, are such that we cannot infer the safety of any practice followed in the East merely from its long continuance.

Grain, however, may be long kept in the open air in climates much less dry than that of Egypt, without injury, except to the superficial layers; for moisture does not penetrate to a great depth in a heap of grain once well dried and kept well aired. When Louis IX. was making his preparations for his campaign in the East, he had large quantities of wine and grain purchased in the Island of Cyprus, and stored up for two years to await his arrival. "When we were come to Cyprus," says Joinville, Histoire de Saint Louis, Section 72, 73, "we found there greate foison of the Kynge's purveyance. . . The wheate and the barley they had piled up in greate heapes in the feeldes, and to looke vpon, they were like vnto mountaynes; for the raine, the whyche hadde beaten vpon the wheate now a longe whyle, had made it to sproute on the toppe, so that it seemed as greene grasse. And whanne they were mynded to carrie it to Egypte, they brake that sod of greene herbe, and dyd finde under the same the wheate and the barley, as freshe as yf menne hadde but nowe thrashed it."]

Total Climatic Influence of the Forest.

Aside from the question of local disturbances and their compensations, it does not seem probable that the forests sensibly affect the general mean of atmospheric temperature of the globe, or the total quantity of precipitation, or even that they had this influence when their extent was vastly greater than at present. The waters cover about three-fourths of the face of the earth, and if we deduct the frozen zones, the peaks and crests of lofty mountains and their craggy slopes, the Sahara and other great African and Asiatic deserts, and all such other portions of the solid surface as are permanently unfit for the growth of wood, we shall find that probably not one-tenth of the total superficies of our planet was ever, at any one time in the present geological period, covered with forests. Besides this, the distribution of forest land, of desert, and of water, is such as to reduce the possible influence of the woods to a low expression; for the forests are, in large proportion, situated in cold or temperate climates, where the action of the sun is comparatively feeble both in elevating temperature and in promoting evaporation; while, in the torrid zone, the desert and the sea—the latter of which always presents an evaporable surface—enormously preponderate. It is, upon the whole, not probable that so small an extent of forest, so situated, could produce a sensible influence on the general climate of the globe, though it might appreciably affect the local action of all climatic elements. The total annual amount of solar heat absorbed and radiated by the earth, and the sum of terrestrial evaporation and atmospheric precipitation, must be supposed constant; but the distribution of heat and of humidity is exposed to disturbance in both time and place by a multitude of local causes, among which the presence or absence of the forest is doubtless one.

So far as we are able to sum up the results, it would appear that, in countries in the temperate zone still chiefly covered with wood, the summers would be cooler, moister, shorter, the winters milder, drier, longer, than in the same regions after the removal of the forest, and that the condensation and precipitation of atmospheric moisture would be, if not greater in total quantity, more frequent and less violent in discharge. The slender historical evidence we possess seems to point to the same conclusion, though there is some conflict of testimony and of opinion on this point.

Among the many causes which, as we have seen, tend to influence the general result, the mechanical action of the forest, if not more important, is certainly more obvious and direct than the immediate effects of its organic processes. The felling of the woods involves the sacrifice of a valuable protection against the violence of chilling winds and the loss of the shelter afforded to the ground by the thick coating of leaves which the forest sheds upon it and by the snow which the woods prevent from blowing away, or from melting in the brief thaws of winter. I have already remarked that bare ground freezes much deeper than that which is covered by beds of leaves, and when the earth is thickly coated with snow, the strata frozen before it fell begin to thaw. It is not uncommon to find the ground in the woods, where the snow lies two or three feet deep, entirely free from frost, when the atmospheric temperature has been for several weeks below the freezing-point, and for some days even below the zero of Fahrenheit. When the ground is cleared and brought under cultivation, the leaves are ploughed into the soil and decomposed, and the snow, especially upon knolls and eminences, is blown off, or perhaps half thawed, several times during the winter. The water from the melting snow runs into the depressions, and when, after a day or two of warm sunshine or tepid rain, the cold returns, it is consolidated to ice, and the bared ridges and swells of earth are deeply frozen. [Footnote: I have seen, in Northern New England, the surface of the open ground frozen to the depth of twenty-two inches, in the month of November, when in the forest-earth no frost was discoverable; and later in the winter, I have known an exposed sand-knoll to remain frozen six feet deep, after the ground in the woods was completely thawed.] It requires many days of mild weather to raise the temperature of soil in this condition, and of the air in contact with it, to that of the earth in the forests of the same climatic region. Flora is already plaiting her sylvan wreath before the corn-flowers which are to deck the garland of Ceres have waked from their winter's sleep; and it is probably not a popular error to believe that, where man has substituted his artificial crops for the spontaneous harvest of nature, spring delays her coming. [Footnote: The conclusion arrived at by Noah Webster, in his very learned and able paper on the supposed change in the temperature of winter, read before the Connecticut Academy of Arts and Sciences in 1799, was as follows: "From a careful comparison of these facts, it appears that the weather, in modern winters, in the United States, is more inconstant than when the earth was covered with woods, at the first settlement of Europeans in the country; that the warm weather of autumn extends further into the winter months, and the cold weather of winter and spring encroaches upon the summer; that, the wind being more variable, snow is less permanent, and perhaps the same remark may be applicable to the ice of the rivers. These effects seem to result necessarily from the greater quantity of heat accumulated in the earth in summer since the ground has been cleared of wood and exposed to the rays of the sun, and to the greater depth of frost in the earth in winter by the exposure of its uncovered surface to the cold atmosphere."—Collection of Papers by Noah Webster, p. 162.]

There are, in the constitution and action of the forest, many forces, organic and inorganic, which unquestionably tend powerfully to produce meteorological effects, and it may, therefore, be assumed as certain that they must and do produce such effects, UNLESS they compensate and balance each other, and herein lies the difficulty of solving the question. To some of these elements late observations give a new importance. For example, the exhalation of aqueous vapor by plants is now believed to be much greater, and the absorption of aqueous vapor by them much less, than was formerly supposed, and Tyndall's views on the relations of vapor to atmospheric heat give immense value to this factor in the problem. In like manner the low temperature of the surface of snow and the comparatively high temperature of its lower strata, and its consequent action on the soil beneath, and the great condensation of moisture by snow, are facts which seem to show that the forest, by protecting great surfaces of snow from melting, must inevitably exercise a great climatic influence. If to these influences we add the mechanical action of the woods in obstructing currents of wind, and diminishing the evaporation and refrigeration which such currents produce, we have an accumulation of forces which MUST manifest great climatic effects, unless—which is not proved and cannot be presumed—they neutralize each other. These are points hitherto little considered in the discussion, and it seems difficult to deny that as a question of ARGUMENT, the probabilities are strongly in favor of the meteorological influence of the woods. The EVIDENCE, indeed, is not satisfactory, or, to speak more accurately, it is non-existent, for there really is next to no trustworthy proof on the subject, but it appears to me a case where the burden of proof must be taken by those who maintain that, as a meteorological agent, the forest is inert.

The question of a change in the climate of the Northern American States is examined in the able Meteorological Report of Mr. Draper, Director of the New York Central Park Observatory, for 1871. The result arrived at by Mr. Draper is, that there is no satisfactory evidence of a diminution in the rainfall, or of any other climatic change in the winter season, in consequence of clearing of the forests or other human action. The proof from meteorological registers is certainly insufficient to establish the fact of a change of climate, but, on the other hand, it is equally insufficient to establish the contrary. Meteorological stations are too few, their observations, in many cases, extend over a very short period, and, for reasons I have already given, the great majority of their records are entitled to little or no confidence. [Footnote: Since these pages were written, the subject of forest meteorology has received the most important contribution ever made to it, in several series of observations at numerous stations in Bavaria, from the year 1866 to 1871, published by Ebermayer, at Aschaffenburg, in 1873, under the title: Die Physikalischen Einwirkungen des Waldes auf Luft und Boden, und seine Klimatologische und Hygienische Bedeutung. I. Band. So far as observations of only five years' duration can prove anything, the following propositions, not to speak of many collateral and subsidiary conclusions, seem to be established, at least for the localities where the observations were made:

1. The yearly mean temperature of wooded soils, at all depths, is lower than that of open grounds, p. 85.

This conclusion, it may be remarked, is of doubtful applicability in regions of excessive climate like the Northern United States and Canada, where the snow keeps the temperature of the soil in the forest above the freezing-point, for a large part and sometimes the whole of the winter, while in unwooded ground the earth remains deeply frozen.

2. The yearly mean atmospheric temperature, other things being equal, is lower in the forest than in cleared grounds, p. 84.

3. Climates become excessive in consequence of extensive clearings, p. 117.

4. The ABSOLUTE humidity of the air in the forest is about the same as in open ground, while the RELATIVE humidity is greater in the former than in the latter case, on account of the lower temperature of the atmosphere in the wood, p. 150.

5. The evaporation from an exposed surface of water in the forest is sixty-four per cent. less than in unwooded grounds, pp. 159,161.

6. About twenty-six per cent. of the precipitation is interrupted and prevented from reaching the ground by the foliage and branches of forest trees, p. 194.

7. In the interior of thick woods, the evaporation from water and from earth is much less than the precipitation, p. 210.

8. The loss of the water of precipitation intercepted by the trees in the forest is compensated by the smaller evaporation from the ground, p. 219.

9. In elevated regions and during the summer half of the year, woods tend to increase the precipitation, p. 202.]

Influence of the Forest on the Humidity of the Soil.

I have hitherto confined myself to the influence of the forest on meteorological conditions, a subject, as has been seen, full of difficulty and uncertainty. Its comparative effects on the temperature, the humidity, the texture and consistence, the configuration and distribution of the mould or arable soil, and, very often, of the mineral strata below, and on the permanence and regularity of springs and greater superficial water-courses, are much less disputable as well as more easily estimated and more important, than its possible value as a cause of strictly climatic equilibrium or disturbance.

The action of the forest on the earth is chiefly mechanical, but the organic process of absorption of moisture by its roots affects the quantity of water contained in the vegetable mould and in the mineral strata near the surface, and, consequently, the consistency of the soil. In treating of the effects of trees on the moisture of the atmosphere, I have said that the forest, by interposing a canopy between the sky and the ground, and by covering the surface with a thick mantle of fallen leaves, at once obstructed insulation and prevented the radiation of heat from the earth. These influences go far to balance each other; but familiar observation shows that, in summer, the forest-soil is not raised to so high a temperature as open grounds exposed to irradiation. For this reason, and in consequence of the mechanical resistance opposed by the bed of dead leaves to the escape of moisture, we should expect that, except after recent rains, the superficial strata of woodland-soil would be more humid than that of cleared land. This agrees with experience. The soil of the natural forest is always moist, except in the extremest droughts, and it is exceedingly rare that a primitive wood suffers from want of humidity. How far this accumulation of water affects the condition of neighboring grounds by lateral infiltration, we do not know, but we shall see, in a subsequent chapter, that water is conveyed to great distances by this process, and we may hence infer that the influence in question is an important one.

It is undoubtedly true that loose soils, stripped of vegetation and broken up by the plough or other processes of cultivation, may, until again carpeted by grasses or other plants, absorb more rain and snow-water than when they were covered by a natural growth; but it is also true that the evaporation from such soils is augmented in a still greater proportion. Rain scarcely penetrates beneath the sod of grass-ground, but runs off over the surface; and after the heaviest showers a ploughed field will often be dried by evaporation before the water can be carried off by infiltration, while the soil of a neighboring grove will remain half saturated for weeks together. Sandy soils frequently rest on a tenacious subsoil, at a moderate depth, as is usually seen in the pine plains of the United States, where pools of rain-water collect in slight depressions on the surface of earth the upper stratum of which is as porous as a sponge. In the open grounds such pools are very soon dried up by the sun and wind; in the woods they remain unevaporated long enough for the water to diffuse itself laterally until it finds, in the subsoil, crevices through which it may escape, or slopes which it may follow to their outcrop or descend along them to lower strata.

Drainage by Roots of Trees.

Becquerel notices a special function of the forest to which I have already alluded, but to which sufficient importance has not, until very recently, been generally ascribed. I refer to the mechanical action of the roots as conductors of the superfluous humidity of the superficial earth to lower strata. The roots of trees often penetrate through subsoil almost impervious to water, and in such cases the moisture, which would otherwise remain above the subsoil and convert the surface-earth into a bog, follows the roots downwards and escapes into more porous strata or is received by subterranean canals or reservoirs. [Footnote: "The roots of vegetables," says d'Hericourt, "perform the office of draining in a manner analogous to that artificially practised in parts of Holland and the British islands. This method consists in driving deeply down into the soil several hundred stakes to the acre; the water filters down along the stakes, and in some cases as favorable results have been obtained by this means as by horizontal drains."-Annales Forestieres, 1837, p. 312.] When the forest is felled, the roots perish and decay, the orifices opened by them are soon obstructed, and the water, after having saturated the vegetable earth, stagnates on the surface and transforms it into ponds and morasses. Thus in La Brenne, a tract of 200,000 acres resting on an impermeable subsoil of argillaceous earth, which ten centuries ago was covered with forests interspersed with fertile and salubrious meadows and pastures, has been converted, by the destruction of the woods, into a vast expanse of pestilential pools and marshes. In Sologne the same cause has withdrawn from cultivation and human inhabitation not less than 1,100,000 acres of ground once well wooded, well drained, and productive.

It is an important observation that the desiccating action of trees, by way of drainage or external conduction by the roots, is greater in the artificial than in the natural wood, and hence that the surface of the ground in the former is not characterized by that approach to a state of saturation which it so generally manifests in the latter. In the spontaneous wood, the leaves, fruits, bark, branches, and dead trunks, by their decayed material and by the conversion of rock into loose earth through the solvent power of the gases they develop in decomposition, cover the ground with an easily penetrable stratum of mixed vegetable and mineral matter extremely favorable to the growth of trees, and at the same time too retentive of moisture to part with it readily to the capillary attraction of the roots.

The trees, finding abundant nutriment near the surface, and so sheltered against the action of the wind by each other as not to need the support of deep and firmly fixed stays, send their roots but a moderate distance downwards, and indeed often spread them out like a horizontal network almost on the surface of the ground. In the artificial wood, on the contrary, the spaces between the trees are greater; they are obliged to send their roots deeper both for mechanical support and in search of nutriment, and they consequently serve much more effectually as conduits for perpendicular drainage.

It is only under special circumstances, however, that this function of the forest is so essential a conservative agent as in the two cases just cited. In a champaign region insufficiently provided with natural channels for the discharge of the waters, and with a subsoil which, though penetrable by the roots of trees, is otherwise impervious to water, it is of cardinal importance; but though trees everywhere tend to carry off the moisture of the superficial strata by this mode of conduction, yet the precise condition of soil which I have described is not of sufficiently frequent occurrence to have drawn much attention to this office of the wood. In fact, in most soils, there are counteracting influences which neutralize, more or less effectually, the desiccative action of roots, and in general it is as true as it was in Seneca's time, that "the shadiest grounds are the moistest." [Footnote: Seneca, Questiones Naturales, iii. 11, 2.]

It is always observed in the American States, that clearing the ground not only causes running springs to disappear, but dries up the stagnant pools and the spongy soils of the low grounds. The first roads in those States ran along the ridges, when practicable, because there only was the earth dry enough to allow of their construction, and, for the same reason, the cabins of the first settlers were perched upon the hills. As the forests have been from time to time removed, and the face of the earth laid open to the air and sun, the moisture has been evaporated, and the removal of the highways and of human habitations from the bleak hills to the sheltered valleys, is one of the most agreeable among the many improvements which later generations have witnessed in the interior of the Northern States. [Footnote: The Tuscan poet Ginati, who hod certainly had little opportunity of observing primitive conditions of nature and of man, was aware that such must have been the course of things in new countries. "You know," says he in a letter to a friend, "that the hills were first occupied by man, because stagnant waters, and afterwards continual wars, excluded men from the plains. But when tranquillity was established and means provided for the discharge of the waters, the low grounds were soon covered with human habitations."— Letters, Firenze, 1864, p. 98.]

Recent observers in France affirm that evergreen trees exercise a special desiccating action on the soil, and cases are cited where large tracts of land lately planted with pines have been almost completely drained of moisture by some unknown action of the trees. It is argued that the alleged drainage is not due to the conducting power of the roots, inasmuch as the roots of the pine do not descend lower than those of the oak and other deciduous trees which produce no such effect, and it is suggested that the foliage of the pine continues to exhale through the winter a sufficient quantity of moisture to account for the drying up of the soil. This explanation is improbable, and I know nothing in American experience of the forest which accords with the alleged facts. It is true that the pines, the firs, the hemlock, and all the spike-leaved evergreens prefer a dry soil, but it has not been observed that such soils become less dry after the felling of their trees. The cedars and other trees of allied families grow naturally in moist ground, and the white cedar of the Northern States, Thuya occidentalis, is chiefly found in swamps. The roots of this tree do not penetrate deeply into the earth, but are spread out near the surface, and of course do not carry off the waters of the swamp by perpendicular conduction. On the contrary, by their shade, the trees prevent the evaporation of the superficial water; but when the cedars are felled, the swamp—which sometimes rather resembles a pool filled with aquatic trees than a grove upon solid ground—often dries up so completely as to be fit for cultivation without any other artificial drainage than, in the ordinary course of cultivation, is given to other new soils. [Footnote: A special dessicative influence has long been ascribed to the maritime pine, which has been extensively planted on the dunes and sand-plains of western France, and it is well established that, under certain conditions, all trees, whether evergreen or deciduous, exercise this function, but there is no convincing proof that in the cases now referred to there is any difference in the mode of action of the two classes of trees. An article by D'Arbois de Jubainville in the Revue des Eaux et Forets for April, 1869, ascribing the same action to the Pinus sylvestris, has excited much attention in Europe, and the facts stated by this writer constitute the strongest evidence known to me in support of the alleged influence of evergreen trees, as distinguished from the draining by downward conduction, which is a function exercised by all trees, under ordinary circumstances, in proportion to their penetration of a bibulous subsoil by tap or other descending roots. The question has been ably discussed by Beraud in the Revue des Deux Mondes for April, 1870, the result being that the drying of the soil by pines is due simply to conduction by the roots, whatever may be the foliage of the tree. See post: Influence of the Forest on Flow of Springs. It is however certain, I believe, that evergreens exhale more moisture in winter than leafless deciduous trees, and consequently some weight is to be ascribed to this element.]

The Forest in Winter.

The influence of the woods on the flow of springs, and consequently on the supply for the larger water-courses, naturally connects itself with the general question of the action of the forest on the humidity of the ground. But the special condition of the woodlands, as affected by snow and frost in the winter of excessive climates, like that of the United States, has not been so much studied as it deserves; and as it has a most important bearing on the superficial hydrology of the earth, I shall make some observations upon it before I proceed to the direct discussion of the influence of the forest on the flow of springs.

To estimate rightly the importance of the forest in our climate as a natural apparatus for accumulating the water that falls upon the surface and transmitting it to the subjacent strata, we must compare the condition and properties of its soil with those of cleared and cultivated earth, and examine the consequently different action of these soils at different seasons of the year. The disparity between them is greatest in climates where, as in the Northern American States and in the extreme North of Europe, the open ground freezes and remains impervious to water during a considerable part of the winter; though, even in climates where the earth does not freeze at all, the woods have still an important influence of the same character. The difference is yet greater in countries which have regular wet and dry seasons, rain being very frequent in the former period, while, in the latter, it scarcely occurs at all. These countries lie chiefly in or near the tropics, but they are not wanting in higher latitudes; for a large part of Asiatic and even of European Turkey is almost wholly deprived of summer rains. In the principal regions occupied by European cultivation, and where alone the questions discussed in this volume are recognized as having, at present, any practical importance, more or less rain falls at all seasons, and it is to these regions that, on this point as well as others, I chiefly confine my attention.

Importance of Snow.

Recent observations in Switzerland give a new importance to the hygrometrical functions of snow, and of course to the forest as its accumulator and protector. I refer to statements of the condensation of atmospheric vapor by the snows and glaciers of the Rhone basin, where it is estimated to be nearly equal to the entire precipitation of the valley. Whenever the humidity of the atmosphere in contact with snow is above the point of saturation at the temperature to which the air is cooled by such contact, the superfluous moisture is absorbed by the snow or condensed and frozen upon its surface, and of course adds so much to the winter supply of water received from the snow by the ground. This quantity, in all probability, much exceeds the loss by evaporation, for during the period when the ground is covered with snow, the proportion of clear dry weather favorable to evaporation is less than that of humid days with an atmosphere in a condition to yield up its moisture to any bibulous substance cold enough to condense it. [Footnote: The hard snow-crust, which in the early spring is a source of such keen enjoyment to the children and youth of the North—and to many older persons in whom the love of nature has kept awake a relish for the simple pleasures of rural life—is doubtless due to the congelation of the vapor condensed by the snow rather than to the thawing and freezing of the superficial stratum; for when the surface is melted by the sun, the water is taken up by the absorbent mass beneath before the temperature falls low enough to freeze it.]

In our Northern States, irregular as is the climate, the first autumnal snows pretty constantly fall before the ground is frozen at all, or when the frost extends at most to the depth of only a few inches. [Footnote: The hard autumnal frosts are usually preceded by heavy rains which thoroughly moisten the soil, and it is a common saying in the North that "the ground will not freeze till the swamps are full."] In the woods, especially those situated upon the elevated ridges which supply the natural irrigation of the soil and feed the perennial fountains and streams, the ground remains covered with snow during the winter; for the trees protect the snow from blowing from the general surface into the depressions, and new accessions are received before the covering deposited by the first fall is melted. Snow is of a color unfavorable for radiation, but, even when it is of considerable thickness, it is not wholly impervious to the rays of the sun, and for this reason, as well as from the warmth of lower strata, the frozen crust of the soil, if one has been formed, is soon thawed, and does not again fall below the freezing-point during the winter. [Footnote: Dr. Williams, of Vermont, made some observations on the comparative temperature of the soil in open and in wooded ground In the years 1789 and 1791, but they generally belonged to the warmer months, and I do not know that any extensive series of comparisons between the temperature of the ground in the woods and in the fields has been attempted in America. Dr. Williams's thermometer was sunk to the depth of ten inches, and gave the following results:

—— | Temperature | Temperature | Time. | of ground in| of ground in| Difference. | pasture. | woods. | —— May 23………………….| 52 | 46 | 6 " 28………………….| 57 | 48 | 9

June 15………………….| 64 | 51 | 13 " 27………………….| 62 | 51 | 11 July 16………………….| 62 | 51 | 11 " 30………………….| 65 1/2 | 55 1/2 | 10 Aug. 15………………….| 68 | 58 | 10 " 31………………….| 59 1/2 | 55 | 4 1/2 Sept.15………………….| 59 1/2 | 55 | 4 1/2 Oct. 1………………….| 59 1/2 | 55 | 4 1/2 " 15………………….| 49 | 49 | 0 Nov. 1………………….| 43 | 43 | 0 " 16………………….| 43 1/2 | 43 1/2 | 0

On the 14th of January, 1791, in a winter remarkable for its extreme severity, he found the ground, on a plain open field where the snow had been blown away, frozen to the depth of three feet and five inches; in the woods where the snow was three feet deep, and where the soil had frozen to the depth of six inches before the snow fell, the thermometer, at six inches below the surface of the ground, stood at 39 degrees. In consequence of the covering of the snow, therefore, the previously frozen ground had been thawed and raised to seven degrees above the freezing-point.—William's Vermont, i., p. 74.

Boussingault's observations are important. Employing three thermometers, one with the bulb an inch below the surface of powdery snow; one on the surface of the ground beneath the snow, then four inches deep; and one in the open air, forty feet above the ground, on the north side of a building, he found, at 5 P.M., the FIRST thermometer at -1.5 degrees Centigrade, the second at 0 degrees, and the THIRD at + 2.5 degrees; at 7 A.M. the next morning, the first stood at -12 degrees, the second at -3.5 degrees and the third at -3 degrees; at 5.30 the same evening No. 1 stood at -1.4 degrees, No. 2 at 0 degrees, and No. 3 at + 3 degrees. Other experiments were tried, and though the temperature was affected by the radiation, which varied with the hour of the day and the state of the sky, the upper surface of the snow was uniformly colder than the lower, or than the open air.

According to the Report of the Department of Agriculture for May and June, 1872, Mr. C. G. Prindle, of Vermont, in the preceding winter, found, for four successive days, the temperature immediately above the snow at 13 degrees below zero; beneath the snow, which was but four inches deep, at 19 degrees above zero; and under a drift two feet deep, at 27 degrees above.

On the borders and in the glades of the American forest, violets and other small plants begin to vegetate as soon as the snow has thawed the soil around their roots, and they are not unfrequently found in full flower under two or three feet of snow.—American Naturalist, May, 1869, pp. 155, 156.

In very cold weather, when the ground is covered with light snow, flocks of the grouse of the Eastern States often plunge into the snow about sunset, and pass the night in this warm shelter. If the weather moderates before morning, a frozen crust is sometimes formed on the surface too strong to be broken by the birds, which consequently perish.] The snow in contact with the earth now begins to melt, with greater or less rapidity, according to the relative temperature of the earth and the air, while the water resulting from its dissolution is imbibed by the vegetable mould, and carried off by infiltration so fast that both the snow and the layers of leaves in contact with it often seem comparatively dry, when, in fact, the under-surface of the former is in a state of perpetual thaw. No doubt a certain proportion of the snow is given off to the atmosphere by direct evaporation, but in the woods, the protection against the sun by even leafless trees prevents much loss in this way, and besides, the snow receives much moisture from the air by absorption and condensation. Very little water runs off in the winter by superficial water-courses, except in rare cases of sudden thaw, and there can be no question that much the greater part of the snow deposited in the forest is slowly melted and absorbed by the earth.

The immense importance of the forest, as a reservoir of this stock of moisture, becomes apparent, when we consider that a large proportion of the summer rain either flows into the valleys and the rivers, because it falls faster than the ground can imbibe it; or, if absorbed by the warm superficial strata, is evaporated from them without sinking deep enough to reach wells and springs, which, of course, depend much on winter rains and snows for their entire supply. This observation, though specially true of cleared and cultivated grounds, is not wholly inapplicable to the forest, particularly when, as is too often the case in Europe, the underwood and the decaying leaves are removed.

The quantity of snow that falls in extensive forests, far from the open country, has seldom been ascertained by direct observation, because there are few meteorological stations in or near the forest. According to Thompson, [Footnote: Thompson's Vermont, Appendix, p. 8.] the proportion of water which falls in snow in the Northern States does not exceed one-fifth of the total precipitation, but the moisture derived from it is doubtless considerably increased by the atmospheric vapor absorbed by it, or condensed and frozen on its surface. I think I can say from experience—and I am confirmed in this opinion by the testimony of competent observers whose attention has been directed specially to the point—that though much snow is intercepted by the trees, and the quantity on the ground in the woods is consequently less than in open land in the first part of the winter, yet most of what reaches the ground at that season remains under the protection of the wood until melted, and as it occasionally receives new supplies the depth of snow in the forest in the latter half of winter is considerably greater than in the cleared fields. Careful measurements in a snowy region in New England, in the month of February, gave a mean of 38 inches in the open ground and 44 inches in the woods. [Footnote: As the loss of snow by evaporation has been probably exaggerated by popular opinion, an observation or two on the subject may not be amiss in this place. It is true that in the open grounds, in clear weather and with a dry atmosphere, snow and ice are evaporated with great rapidity even when the thermometer is much below the freezing-point; and Darwin informs us that the snow on the summit of Aconcagua, 23,000 feet high, and of course in a temperature of perpetual frost, is sometimes carried off by evaporation. The surface of the snow in our woods, however, does not indicate much loss in this way. Very small deposits of snow-flakes remain unevaporated in the forest, for many days after snow which fell at the same time in the cleared field has disappeared without either a thaw to melt it or a wind powerful enough to drift it away. Even when bared of their leaven, the trees of a wood obstruct, in an important degree, both the direct action of the sun's rays on the snow and the movement of drying and thawing winds.

Dr. Piper (Trees of America, p. 48) records the following observations: "A body of snow, one foot in depth and sixteen feet square, was protected from the wind by a tight board fence about five feet high, while another body of snow, much more sheltered from the sun than the first, six feet in depth, and about sixteen feet square, was fully exposed to the wind. When the thaw came on, which lasted about a fortnight, the larger body of snow was entirely dissolved in less than a week, while the smaller body was not wholly gone at the end of the second week. "Equal quantities of snow were placed in vessels of the samekind and capacity, the temperature of the air being seventy degrees. In the one case, a constant current of air was kept passing over the open vessel, while the other was protected by a cover. The snow in the first was dissolved in sixteen minutes, while the latter had a small unthawed proportion remaining at the end of eighty-five minutes." The snow in the woods is protected in the same way, though not literally to the same extent, as by the fence in one of these cases and the cover in the other.]

The general effect of the forest in cold climates is to assimilate the winter state of the ground to that of wooded regions under softer skies; and it is a circumstance well worth noting, that in Southern Europe, where Nature has denied to the earth a warm winter-garment of flocculent snow, she has, by one of those compensations in which her empire is so rich, clothed the hillsides with umbrella and other pines, ilexes, cork-oaks, bays and other trees of persistent foliage, whose evergreen leaves afford to the soil a protection analogous to that which it derives from snow in more northern climates.

The water imbibed by the soil in winter sinks until it meets a more or less impermeable or a saturated stratum, and then, by unseen conduits, slowly finds its way to the channels springs, or oozes out of the ground in drops which unite in rills, and so all is conveyed to the larger streams, and by them finally to the sea. The water, in percolating through the vegetable and mineral layers, acquires their temperature, and is chemically affected by their action, but it carries very little matter in mechanical suspension.

The process I have described is a slow one, and the supply of moisture derived from the snow, augmented by the rains of the following seasons, keeps the forest-ground, where the surface is level or but moderately inclined, in a state of approximate saturation throughout almost the whole year. [Footnote: The statements I have made, here and elsewhere, respecting the humidity of the soil in natural forests, have been, I understand, denied by Mr. T. Meehan, a distinguished American naturalist, in a paper which I have not seen He is quoted as maintaining, among other highly questionable propositions that no ground is "so dry in its subsoil as that which sustains a forest on its surface." In open, artificially planted woods, with a smooth and regular surface, and especially in forests where the fallen leaves and branches are annually burnt or carried off, both the superficial and the subjacent strata may under certain circumstances, become dry, but this rarely, if ever, happens in a wood of spontaneous growth, undeprived of the protection afforded by its own droppings, and of the natural accidents of surface which tend to the retention of water. See, on this point, a very able article by Mr. Henry Stewart, in the New York Tribune of November 23, 1873.] It may be proper to observe here that in Italy, and in many parts of Spain and France, the Alps, the Apennines, and the Pyrenees, not to speak of less important mountains, perform the functions which provident nature has in other regions assigned to the forest, that is, they act as reservoirs wherein is accumulated in winter a supply of moisture to nourish the parched plains during the droughts of summer. Hence, however enormous may be the evils which have accrued to the above-mentioned countries from the destruction of the woods, the absolute desolation which would otherwise have smitten them through the folly of man, has been partially prevented by those natural dispositions, by means of which there are stored up in the glaciers, in the snow-fields, and in the basins of mountains and valleys, vast deposits of condensed moisture which are afterwards distributed in a liquid form during the season in which the atmosphere furnishes a slender supply of the beneficent fluid so indispensable to vegetable and animal life. [Footnote: The accumulation of snow and ice upon the Alps and other mountains—which often fills up valleys to the height of hundreds of feet—is due not only to the fall or congealed and crystallized vapor in the form of snow, to the condensation of atmospheric vapor on the surface of snow-fields and glaciers, and to a temperature which prevents the rapid melting of snow, but also to the well-known fact that, at least up to the height of 10,000 feet, rain and snow are more abundant on the mountains than at lower levels.

But another reason may be suggested for the increase of atmospheric humidity, and consequently of the precipitation of aqueous vapor on mountain chains. In discussing the influence of mountains on precipitation, meteorologists have generally treated the popular belief, that mountains "attract" to them clouds floating within a certain distance from them, as an ignorant prejudice, and they ascribe the appearance of clouds about high peaks solely to the condensation of the humidity of the air carried by atmospheric currents up the slopes of the mountain to a colder temperature. But if mountains do not really draw clouds and invisible vapors to them, they are an exception to the universal law of attraction. The attraction of the small Mount Shehallien was found sufficient to deflect from the perpendicular, by a measurable quantity, a plummet weighing but a few ounces. Why, then, should not greater masses attract to them volumes of vapor weighing many tons, and floating freely in the atmosphere within moderate distances of the mountains ]

Summer Rains, Importance of.

Babinet quotes a French proverb: "Summer rain wets nothing," and explains it by saying that at that season the rainwater is "almost entirely carried off by evaporation." "The rains of summer," he adds, "however abundant they may be, do not penetrate the soil beyond the depth of six or eight inches. In summer the evaporating power of the heat is five or six times greater than in winter, and this force is exerted by an atmosphere capable of containing five or six times as much vapor as in winter." "A stratum of snow which prevents evaporation [from the ground], causes almost all the water that composes it to filter into the earth, and forms a provision for fountains, wells, and streams which could not be furnished by any quantity whatever of summer rain. This latter, useful to vegetation like the dew, neither penetrates the soil nor accumulates a store to supply the springs and to be given out again into the open air." [Footnote: Etudes et Lectures, vol. vi., p. 118. The experiments or Johnstrup in the vicinity of Copenhagen, where the mean annual precipitation is 23 1/2 inches, and where the evaporation must be less than in the warmer and drier atmosphere of France, form the most careful series of observations on this subject which I have met with. Johnstrup found that at the depth at a metre and a half (50 inches) the effects of rain and evaporation were almost imperceptible, and became completely so at a depth of from two to three metres (6 1/2 to 10 feet). During the summer half of the year the evaporation rather exceeded the rainfall; during the winter half the entire precipitation was absorbed by the soil and transmitted to lower strata by infiltration. The stratum between one metre and a half (50 inches) and three metres (10 feet) from the surface was then permanently in the condition of a saturated sponge, neither receiving nor losing humidity during the summer half of the year, but receiving from superior, and giving off to lower, strata an equal amount of moisture during the winter half.—Johnstrup, Om Fugtighedens Bezagelse i den naturlige Jordbund. Kjobenhavn, 1866.]

This conclusion, however applicable to the climate and to the soil of France, is too broadly stated to be received as a general truth; and in countries like the United States, where rain is comparatively rare during the winter and abundant during the summer half of the year, common observation shows that the quantity of water furnished by deep wells and by natural springs depends almost as much upon the rains of summer as upon those of the rest of the year, and consequently that a large portion of the rain of that season must find its way into strata too deep for the water to be wasted by evaporation.

[Footnote: According to observations at one hundred military stations in the United States, the precipitation ranges from three and a quarter inches at Fort Yuma in California to about seventy-two inches at Fort Pike, Louisiana, the mean for the entire territory, not including Alaska, being thirty-six inches. In the different sections of the Union it is as follows:

North-eastern States……………… 41 inches, New York………………………… 36 " Middle States……………………. 40 1/2 " Ohio……………………………. 40 " Southern States………………….. 51 " S. W. States and Indian Territories… 39 1/2 " Western States and Territories…….. 30 " Texas and New Mexico……………… 24 1/2 " California………………………. 18 1/2 " Oregon and Washington Territory……. 50 "

The mountainous regions, it appears, do not recieve the greatest amount of precipitation. The avenge downfall of the Southern States bordering on the Atlantic and the Gulf of Mexico exceeds the mean of the whole United States, being no less than fifty-one inches, while on the Pacific coast it ranges from fifty to fifty-six inches.

As a general rule, it may be stated that at the stations on or near the sea-coast the precipitation is greatest in the spring months, though there are several exceptions to this remark, and at a large majority of the stations the downfall is considerably greater in the summer months than at any other season.]

Dalton's experiments in the years 1796, 1797, and 1798 appeared to show that the mean absorption of the downfall by the earth in those years was twenty-nine per cent.

Dickinson, employing the same apparatus for eight years, found the absorption to vary widely in different years, the mean being forty-seven per cent.

Charnock's experiments in two years show an absorption of from seventeen to twenty-seven per cent.] Besides, even admitting that the water from summer rains is so completely evaporated as to contribute nothing directly to the supply of springs, it at least tends indirectly to maintain their flow, because it saturates in part the atmosphere, and at the same time it prevents the heat of the sun from drying the earth to still greater depths, and bringing within the reach of evaporation the moisture of strata which ordinarily do not feel the effects of solar irradiation.

Influence of the Forest on the Flow of Springs.

It is an almost universal and, I believe, well-founded opinion, that the protection afforded by the forest against the escape of moisture from its soil by superficial flow and evaporation insures the permanence and regularity of natural springs, not only within the limits of the wood, but at some distance beyond its borders, and thus contributes to the supply of an element essential to both vegetable and animal life. As the forests are destroyed, the springs which flowed from the woods, and, consequently, the greater water-courses fed by them, diminish both in number and in volume. This fact is so familiar throughout the American States and the British Provinces, that there are few old residents of the interior of those districts who are not able to testify to its truth as a matter of personal observation. My own recollection suggests to me many instances of this sort, and I remember one case where a small mountain spring, which disappeared soon after the clearing of the ground where it rose, was recovered about twenty years ago, by simply allowing the bushes and young trees to grow up on a rocky knoll, not more than half an acre in extent, immediately above the spring. The ground was hardly shaded before the water reappeared, and it has ever since continued to flow without interruption. The hills in the Atlantic States formerly abounded in springs and brooks, but in many parts of these States which were cleared a generation or two ago, the hill-pastures now suffer severely from drought, and in dry seasons furnish to cattle neither grass nor water.

Almost every treatise on the economy of the forest adduces facts in support of the doctrine that the clearing of the woods tends to diminish the flow of springs and the humidity of the soil, and it might seem unnecessary to bring forward further evidence on this point. [Footnote: "Why go so far for the proof of a phenomenon that is repeated every day under our own eyes, and of which every Parisian may convince himself, without venturing beyond the Bois de Boulogne or the forest of Meudon Let him, after a few rainy days, pass alone the Chevreuse road, which is bordered on the right by the wood, on the left by cultivated fields. The fall of water and the continuance of the rain have been the same on both sides; but the ditch on the side of the forest will remain filled with water proceeding from the infiltration through the wooded soil, long after the other, contiguous to the open ground, has performed its office of drainage and become dry. The ditch on the left will have discharged in a few hours a quantity of water, which the ditch on the right requires several days to receive and carry down to the valley."—Clave, Etudes, etc., pp. 53, 54.] But the subject is of too much practical importance and of too great philosophical interest to be summarily disposed of; and it ought to be noticed that there is at least one case—that of some loose sandy soils which, as observed by Valles, [Footnote: Valles, Etudes sur les Inondations, p. 472.] when bared of wood very rapidly absorb and transmit to lower strata the water they receive from the atmosphere—where the removal of the forest may increase the flow of springs at levels below it, by exposing to the rain and melted snow a surface more bibulous, and at the same time less retentive, than its original covering. Under such circumstances, the water of precipitation, which had formerly been absorbed by the vegetable mould and retained until it was evaporated, might descend through porous earth until it meets an impermeable stratum, and then be conducted along it, until, finally, at the outcropping of this stratum, it bursts from a hillside as a running spring. But such instances are doubtless too rare to form a frequent or an important exception to the general law, because it is very seldom the case that such a soil as has just been supposed is covered by a layer of vegetable earth thick enough to retain, until it is evaporated, all the rain that falls upon it, without imparting any water to the strata below it.

If we look at the point under discussion as purely a question of fact, to be determined by positive evidence and not by argument, the observations of Boussingault are, both in the circumstances they detail and in the weight to be attached to the testimony, among the most important yet recorded. The interest of the question will justify me in giving, nearly in Boussingault's own words, the facts and some of the remarks with which he accompanies the detail of them. "In many localities," he observes, [Footnote: Economie Rurale t. ii, p. 780.] "it has been thought that, within a certain number of years, a sensible diminution has been perceived in the volume of water of streams utilized as a motive-power; at other points, there are grounds for believing that rivers have become shallower, and the increasing breadth of the belt of pebbles along their banks seems to prove the loss of a part of their water; and, finally, abundant springs have almost dried up. These observations have been principally made in valleys bounded by high mountains, and it has been noticed that this diminution of the waters has immediately followed the epoch when the inhabitants have begun to destroy, unsparingly, the woods which were spread over the face of the land. "And here lies the practical point of the question; for if it is once established that clearing diminishes the volume of streams, it is less important to know to what special cause this effect is due. The rivers which rise within the valley of Aragua, having no outlet to the ocean, form, by their union, the Lake of Tacarigua or Valencia, having a length of about two leagues and a half [= 7 English miles].

At the time of Humboldt's visit to the valley of Aragua, the inhabitants were struck by the gradual diminution which the lake had been undergoing for thirty years. In fact, by comparing the descriptions given by historians with its actual condition, even making large allowance for exaggeration, it was easy to see that the level was considerably depressed. The facts spoke for themselves. Oviedo, who, toward the close of the sixteenth century, had often traversed the valley of Aragua, says positively that New Valencia was founded, in 1555, at half a league from the Lake of Tacarigua; in 1800, Humboldt found this city 5,260 metres [= 3 1/2 English miles] from the shore.

"The aspect of the soil furnished new proofs. Many hillocks on the plain retain the name of islands, which they more justly bore when they were surrounded by water. The ground laid bare by the retreat of the lake was converted into admirable plantations; and buildings erected near the lake showed the sinking of the water from year to year. In 1796, new islands made their appearance. A fortress built in 1740 on the island of Cabrera, was now on a peninsula; and, finally, on two granitic islands, those of Cura and Cabo Blanco, Humboldt observed among the shrubs, somo metres above the water, fine sand filled with helicites.

"These clear and positive facts suggested numerous explanations, all assuming a subterranean outlet, which permitted the discharge of the water to the ocean. Humboldt disposed of these hypotheses, and did not hesitate to ascribe the diminution of the waters of the lake to the numerous clearings which had been made in the valley of Aragua within half a century."

Twenty-two years later, Boussingault explored the valley of Aragua. For some years previous, the inhabitants had observed that the waters of the lake were no longer retiring, but, on the contrary, were sensibly rising. Grounds, not long before occupied by plantations, were submerged. The islands of Nuevas Aparecidas, which appeared above the surface in 1796, had again become shoals dangerous to navigation. Cabrera, a tongue of land on the north side of the valley, was so narrow that the least rise of the water completely inundated it. A protracted north wind sufficed to flood the road between Maracay and New Valencia. The fears which the inhabitants of the shores had so long entertained were reversed. Those who had explained the diminution of the lake by the supposition of subterranean channels were suspected of blocking them up, to prove themselves in the right.

During the twenty-two years which had elapsed, the valley of Aragua had been the theatre of bloody struggles, and war had desolated these smiling lands and decimated their population. At the first cry of independence a great number of slaves found their liberty by enlisting under the banners of the new republic; the great plantations were abandoned, and the forest, which in the tropics so rapidly encroaches, had soon recovered a large proportion of the soil which man had wrested from it by more than a century of constant and painful labor.

Boussingault proceeds to state that two lakes near Ubate, in New Granada, had formed but one, a century before his visit; that the waters were gradually retiring, and the plantations extending over the abandoned bed; that, by inquiry of old hunters and by examination of parish records, he found that extensive clearings had been made and were still going on.

He found, also, that the length of the Lake of Fuquene, in the same valley, had, within two centuries, been reduced from ten leagues to one and a half, its breadth from three leagues to one. At the former period, the neighboring mountains were well wooded, but at the time of his visit the mountains had been almost entirely stripped of their wood. Our author adds that other cases, similar to those already detailed, might be cited, and he proceeds to show, by several examples, that the waters of other lakes in the same regions, where the valleys had always been bare of wood, or where the forests had not been disturbed, had undergone no change of level.

Boussingault further states that the lakes of Switzerland have sustained a depression of level since the too prevalent destruction of the woods, and arrives at the general conclusion that, "in countries where great clearings have been made, there has most probably been a diminution in the living waters which flow upon the surface of theground." This conclusion he further supports by two examples: one, where a fine spring, at the foot of a wooded mountain in the Island of Ascension, dried up when the mountain was cleared, but reappeared when the wood was replanted; the other at Marmato, in the province of Popayan, where the streams employed to drive machinery were much diminished in volume, within two years after the clearing of the heights from which they derived their supplies. This latter is an interesting case, because, although the rain-gauges, established as soon as the decrease of water began to excite alarm, showed a greater fall of rain for the second year of observation than the first, yet there was no appreciable increase in the flow of the mill-streams. From these cases, the distinguished physicist infers that very restricted local clearings may diminish and even suppress springs and brooks, without any reduction in the total quantity of rain.

It will have been noticed that these observations, with the exception of the last two cases, do not bear directly upon the question of the diminution of springs by clearings, but they logically infer it from the subsidence of the natural reservoirs which springs once filled. There is, however, no want of positive evidence on this subject. Marchand cites the following instances: "Before the felling of the woods, within the last few years, in the valley of the Soulce, the Combe-es-Monnin and the Little Valley, the Sorne furnished a regular and sufficient supply of water for the ironworks of Unterwyl, which was almost unaffected by drought or by heavy rains. The Sorne has now become a torrent, every shower occasions a flood, and after a few days of fine weather, the current falls so low that it has been necessary to change the water-wheels, because those of the old construction are no longer able to drive the machinery, and at last to introduce a steam-engine to prevent the stoppage of the works for want of water.

"When the factory of St. Ursanne was established, the river that furnished its power was abundant, and had, from time immemorial, sufficed for the machinery of a previous factory. Afterwards, the woods near its sources were cut. The supply of water fell off in consequence, the factory wanted water for half the year, and was at last obliged to stop altogether.

"The spring of Combefoulat, in the commune of Seleate, was well known as one of the best in the country; it was remarkably abundant, and sufficient, in the severest droughts, to supply all the fountains of the town; but as soon as considerable forests were felled in Combe-de-pre Martin and in the valley of Combefoulat, the famous spring, which lies below these woods, has become a mere thread of water, and disappears altogether in times of drought.

"The spring of Varieux, which formerly supplied the castle of Pruntrut, lost more than half its water after the clearing of Varieux and Rougeoles. These woods have been replanted, the young trees are growing well, and, with the woods, the waters of the spring are increasing.

"The Dog Spring between Pruntrut and Bressancourt has entirely vanished since the surrounding forest-grounds were brought under cultivation.

"The Wolf Spring, in the commune of Soubey, furnishes a remarkable example of the influence of the woods upon fountains. A few years ago this spring did not exist. At the place where it now rises, a small thread of water was observed after very long rains, but the stream disappeared with the rain. The spot is in the middle of a very steep pasture inclining to the south. Eighty years ago, the owner of the land, perceiving that young firs were shooting up in the upper part of it, determined to let them grow, and they soon formed a flourishing grove. As soon as they were well grown, a fine spring appeared in place of the occasional rill, and furnished abundant water in the longest droughts. For forty or fifty years this spring was considered the best in the Clos du Doubs. A few years since, the grove was felled, and the ground turned again to a pasture. The spring disappeared with the wood, and is now as dry as it was ninety years ago." [Footnote: Ueber Die Entwaldung Der Gebirge, pp. 20 et seqq.]

Siemoni gives the following remarkable facts from his own personal observation:

"In a rocky nook near the crest of a mountain in the Tuscan Apennines, there flowed a clear, cool, and perennial fountain, uniting three distinct springs in a single current. The ancient beeches around and particularly above the springs were felled. On the disappearance of the wood, the springs ceased to flow, except in a thread of water in rainy weather, greatly inferior in quality to that of the old fountain. The beeches were succeeded by firs, and as soon as they had grown sufficiently to shade the soil, the springs begun again to flow, and they gradually returned to their former abundance and quality." [Footnote: Manuale D'arte Forestale. 2me editione, p. 492.]

This and the next preceding case are of great importance both as to the action of the wood in maintaining springs, and particularly as tending to prove that evergreens do not exercise the desiccative influence ascribed to them in France. The latter instance shows, too, that the protective influence of the wood extends far below the surface, for the quality of the water was determined, no doubt, by the depth from which it was drawn. The slender occasional supply after the beeches were cut was rain-water which soaked through the superficial humus and oozed out at the old orifices, carrying the taste and temperature of the vegetable soil with it; the more abundant and grateful water which flowed before the beeches were cut, and after the firs were well grown, came from a deeper source and had been purified, and cooled to the mean temperature of the locality, by filtering through strata of mineral earth. "The influence of the forest on springs," says Hummel, "is strikingly shown by an instance at Heilbronn. The woods on the hills surrounding the town are cut in regular succession every twentieth year. As the annual cuttings approach a certain point, the springs yield less water, some of them none at all; but as the young growth shoots up, they flow more and more freely, and at length bubble up again in all their original abundance." [Footnote: Physische Geographie, p. 32.] Dr. Piper states the following case: "Within about half a mile of my residence there is a pond upon which mills have been standing for a long time, dating back, I believe, to the first settlement of the town. These have been kept in constant operation until within some twenty or thirty years, when the supply of water began to fail. The pond owes its existence to a stream that has its source in the hills which stretch some miles to the south. Within the time mentioned, these hills, which were clothed with a dense forest, have been almost entirely stripped of trees; and to the wonder and loss of the mill-owners, the water in the pond has failed, except in the season of freshets; and, what was never heard of before, the stream itself has been entirely dry. Within the last ten years a new growth of wood has sprung up on most of the land formerly occupied by the old forest; and now the water runs through the year, notwithstanding the great droughts of the last few years, going back from 1856."

Dr. Piper quotes from a letter of William C. Bryant the following remarks: "It is a common observation that our summers are becoming drier and our streams smaller. Take the Cuyahoga as an illustration. Fifty years ago large barges loaded with goods went up and down that river, and one of the vessels engaged in the battle of Lake Erie, in which the gallant Perry was victorious, was built at Old Portage, six miles north of Albion, and floated down to the lake. Now, in an ordinary stage of the water, a canoe or skiff can hardly pass down the stream. Many a boat of fifty tons burden has been built and loaded in the Tuscarawas, at New Portage, and sailed to New Orleans without breaking bulk. Now, the river hardly affords a supply of water at New Portage for the canal. The same may be said of other streams—they are drying up. And from the same cause—the destruction of our forests—our summers are growing drier and our winters colder." [Footnote: The Trees of America, pp. 50, 51.]

No observer has more carefully studied the influence of the forest upon the flow of the waters, or reasoned more ably on the ascertained phenomena, than Cantegril. The facts presented in the following case, communicated by him to the Ami des Sciences for December, 1859, are as nearly conclusive as any single instance well can be:

"In the territory of the commune of Labruguiere there is a forest of 1,834 hectares [4,530 acres], known by the name of the Forest of Montaut, and belonging to that commune. It extends along thenorthern slope of the Black Mountains. The soil is granitic, the maximum altitude 1,243 metres [4,140 feet], and the inclination ranges between 15 and 60 to 100.

"A small current of water, the brook of Caunan, takes its rise in this forest, and receives the waters of two-thirds of its surface. At the lower extremity of the wood and on the stream are several fulleries, each requiring a force of eight horse-power to drive the water-wheels which work the stampers. The commune of Labruguiere had been for a long time famous for its opposition to forest laws. Trespasses and abuses of the right of pasturage had converted the wood into an immense waste, so that this vast property now scarcely sufficed to pay the expense of protecting it, and to furnish the inhabitants with a meagre supply of fuel. While the forest was thus ruined, and the soil thus bared, the water, after every abundant rain, made an eruption into the valley, bringing down a great quantity of pebbles which still clog the current of the Caunan. The violence of the floods was sometimes such that they were obliged to stop the machinery for some time. During the summer another inconvenience was felt. If the dry weather continued a little longer than usual, the delivery of water became insignificant. Each fullery could for the most part only employ a single set of stampers, and it was not unusual to see the work entirely suspended.

"After 1840, the municipal authority succeeded in enlightening the population as to their true interests. Protected by a more watchful supervision, aided by well-managed replantation, the forest has continued to improve to the present day. In proportion to the restoration of the forest, the condition of the manufactories has become less and less precarious, and the action of the water is completely modified. For example, sudden and violent floods, which formerly made it necessary to stop the machinery, no longer occur. There is no increase in the delivery until six or eight hours after the beginning of the rain; the floods follow a regular progression till they reach their maximum, and decrease in the same manner. Finally, the fulleries are no longer forced to suspend work in summer; the water is always sufficiently abundant to allow the employment of two sets of stampere at least, and often even of three.

"This example is remarkable in this respect, that, all other circumstances having remained the same, the changes in the action of the stream can be attributed only to the restoration of the forest—changes which may be thus summed up: diminution of flood-water during rains—increase of delivery at other seasons."

Becquerel and other European writers adduce numerous other cases where the destruction of forests has caused the disappearance of springs, a diminution in the volume of rivers, and a lowering of the level of lakes, and in fact, the evidence in support of the doctrine I have been maintaining on this subject seems to be as conclusive as the nature of the case admits. [Footnote: See, in the Revue des Eaux et Forets for April, 1867, an article entitled De l'influence des Forets sur le Regime des Eaux, and the papers in previous numbers of the same journal therein referred to.] We cannot, it is true, arrive at the same certainty and precision of result in these inquiries as in those branches of physical research where exact quantitative appreciation is possible, and we must content ourselves with probabilities and approximations. We cannot positively affirm that the precipitation in a given locality is increased by the presence, or lessened by the destruction, of the forest, and from our ignorance of the subterranean circulation of the waters, we cannot predict, with certainty, the drying up of a particular spring as a consequence of the felling of the wood which shelters it; but the general truth, that the flow of springs and the normal volume of rivers rise and fall with the extension and the diminution of the woods where they originate and through which they run, is as well established as any proposition in the science of physical geography. [Footnote: Some years ago it was popularly believed that the volume of the Mississippi, like that of the Volga and other rivers of the Eastern Hemisphere, was diminished by the increased evaporation from its basin and the drying up of the springs in consequence of the felling of the forests in the vicinity of the source of its eastern affluents. The boatmen of this great river and other intelligent observers now assure us, however, that the mean and normal level of the Mississippi has risen within a few years, and that in consequence the river is navigable at low water for boats of greater draught and at higher points in its course than was the case twenty-five years ago. This supposed increase of volume has been attributed by some to the recent re-wooding of the prairies, but the plantations thus far made are not yet sufficiently extensive to produce an appreciable effect of this nature; and besides, while young trees have covered some of the prairies, the destruction of the forest has been continued perhaps in a greater proportion in other parts of the basin of the river. A more plausible opinion is that the substitution of ground that is cultivated, and consequently spongy and absorbent, for the natural soil of the prairies, has furnished a reservoir for the rains which are absorbed by the earth and carried gradually to the river by subterranean flow, instead of running off rapidly from the surface, or, as is more probable, instead of evaporating or being taken up by the vigorous herbaceous vegetation which covers the natural prairie.