soil, and substituting débris and unfertile ground from the mountains, and pauperizing the once productive lands, inflicting damages, which now, by the expenditure of millions of dollars yearly, and with the exercise of the greatest ingenuity, can hardly be repaired. Within the last twenty years, France has reforested about two hundred and fifty thousand acres of mountain-lands, at a cost of $30,000,000 of which the Government paid one half, the local communities the other half. In addition, two hundred thousand acres of sand-dunes, which were the result of injudicious clearing, have been reforested since 1862, and thus been made productive.
An extensive literature on the subject of forest benefits has now accumulated. But I do not intend to rehearse these often-cited arguments, which are so well elaborated in George P. Marsh's classical book, "The Earth as modified by Man." We must admit, however, that historical evidence alone can not be held sufficient proof of any natural law, and a problem of natural science needs for its solution to be subjected to scientific methods of investigation and reasoning. This has only lately been done with regard to forest influences; and, though systematic and continuous observations have not yet extended over a long period, we are already prepared at least to understand the character, though not always the extent, of the part which the forest plays in the economy of Nature. We have learned to discriminate between the different functions of forest influences. We have learned that the mechanical influence of forest-cover upon hydrologic and soil conditions is undeniable; we have learned that climatic changes due to deforestation may be favorable as well as unfavorable; that the great characteristics of a climate, due to cosmic conditions, such as the twofold movement of the earth, the presence of water-surfaces, elevation, the prevailing winds, etc., are probably beyond the reach of forest influence; that such influence must, in the main, be local, and its nature and extent be dependent largely upon the geography of the locality.
The rationale of forest influences is easily enough understood, if we consider them step by step. The climate of a locality—i. e., the interdependent oscillations of temperature and humidity of the air (not as it is popularly expressed the mean condition of these factors)—is, in the first place, dependent upon the heating effect of the sun's rays; in one word, upon "insolation," The temperature of the air derives its heat, for the most part, only from contact with the heated earth or objects on the earth, and by radiation from these. Any mechanical barrier, then, against insolation of the soil, like a shady, dense forest, must have the effect of lowering the temperature of the soil and consequently of the air above it. The immediate consequence of this is diminished evaporation from the surface of the soil; while, on the other hand, the transpiration through the leaves makes the groundwater of greater depths available to the atmosphere. Thus cooler and