The Influence of Forests on Glimaie. 449 



over an area of 40° long, in N. America, found that the mean 

 temperature of a locality is not changed by the destruction of 

 its forests. But though the mean temperature may be un- 

 altered, the distribution of heat at given times may be affected 

 so as to produce a modification of climate, and Humboldt con- 

 sidered that the destruction of trees ought to occasion an aug- 

 mentation of mean temperature by removing several causes of 

 cooling action.* 



Up to the present time sufficient attention has not been 

 paid to the influence of the soil exposed by tree destruction. 

 The temperature of a soil varies according to whether it is dry 

 or moist, calcareous, sandy, or clayey, and experiments show 

 that two soils, one wet and the other dry, exposed to solar 

 radiation, will differ 6° or 7° Cent, when the air temperature is 

 25° Cent, or 77° Fahr. ; and for hnmus (vegetable mould) the 

 difference will sometimes be as much as 12°. The physical 

 character of the soil is likewise important ; land covered with 

 flints does not cool so fast as sand, and flinty soils are on this 

 account found better for vines than chalky or clayey soils, 

 which cool quicker. ., The destruction of trees in a region 

 having a siliceous or silico-calcareous soil produces a maximum 

 elevation of temperature. Thus, western Europe owes the 

 mildness of its climate to warm currents which arrive from the 

 deserts of Sahara, and if terrestrial changes should convert 

 these deserts into forests, its climate would be more severe. 

 In America we find the tropical regions occupied by vast 

 forests, immense savannahs, and great rivers, which cannot 

 produce hot air-currents like the Sahara desert, and soften 

 the climate of the north. Thus, at equal latitudes, North 

 America is colder than Europe, as we see from the direction 

 of isothermal lines, and from its agriculture. 



The influence of forests on the quantity of running water 

 in a country is very important ; but it is extremely difficult to 

 say a priori whether the destruction of part of a forest will 

 augment this or that spring or river. We only know the 

 result after the forest has been cut down. Springs in general 

 are due to infiltrations of rain water, which traverse the soil 

 until they reach an impermeable layer, down which they run, 

 if it be inclined, until they escape as streams or fountains. 

 Such springs are usually found amongst mountains, but forests 

 contribute equally to their formation, not only by means of the 

 moisture which they produce and condense, but likewise on 

 account of the obstacles they impose to evaporation from 

 the soil, and by the action of their roots in rendering soils 

 permeable. 



* In connexion with these inquiries, Professor Tyndall'» researches on the 

 action of atmospheric humidity in arresting radiation should be considered. 



