THE FOREST AND SPRINGS 363 



are generally fed by a series of intercommunicating reservoirs. Sometimes also they 

 have a considerable and remarkably regular flow. This type of spring is very common 

 in the Jura; the springs of Cuisance, Luzon, and a great number of others furnish good 

 examples. The famous fountain of Vaucluse belongs to the same type. 



Space will not permit us to concern ourselves here with this last category of springs, 

 which do not, properly speaking, come under the definition given above, because they 

 are only waste weirs of a lake or a series of subterranean lakes. 



Special Remarks on the Action of Forests on Springs. Do forests favor the 

 feeding of springs? Formerly there was no doubt on this subject. Our fathers always 

 considered forests and springs as co-partners and Buffon wrote, 8 "The more a country 

 is cleared the poorer it becomes in water." The Forest Code of 1827 only sanctions 

 the general belief of the period when it authorizes the forest officials to oppose the 

 clearing of forests with a view to protecting the existence of the springs. 9 One must 

 allow that such an ancient, prevalent, and persistent belief constitutes a strong pre- 

 sumption in favor of the utilitarian purpose of a forest; common opinion can only be 

 the result of proved facts. 



It is only recently that doubts have been expressed on this point. Cases have been 

 quoted where reafforestation has reduced the output of the springs. On the other 

 hand, marshy grounds have been known to lose their superficial water in consequence of 

 reafforestation, and it was concluded from this fact that the forest acted as a kind of 

 pump to inhale in bulk, by means of its roots, the free water of the soil and return it 

 to the air by means of the evaporation of its leaves. It might, however, be objected 

 to in this last case that it is equally admissible that forests have caused superficial 

 stagnant water to disappear by favoring their infiltration. 



It is an extremely difficult and delicate matter to ascertain by direct observation the 

 influence of the afforestation or deforestation of a soil on the output of a spring. It 

 is in fact only by chance in many cases that we are able to recognize exactly the place 

 whence the waters filter which we find oozing forth at a given point; the natural 

 reservoirs may be very far removed 10 from the place where they appear above ground 

 and be separated from them by valleys, heights, etc. The exact defining of the feed- 

 ing basin of a spring is sometimes a very complicated problem, enough to embarrass 

 the most experienced of geological specialists. The direction of the springs, too, is 

 often altered by works such as the cuttings made for roads or railways, for galleries in 

 mines, etc., undertaken very far away from the place where these waters become visible. 

 Certain springs are so superficial that shallow ditches or simple farm drainage can 

 deflect them. Finally the actual output of a spring depends especially on the rainfall 

 of the current year, a factor which is always in the preponderance. A certainty could 

 only exist in the case where reafforestation or clearings practiced on a large scale would 

 have affected to a permanent and notable degree the output of all the springs of a given 

 region. It has often been asserted that this was the case, and was taken to prove that 

 the general level of the waterflow had diminished in many parts, in proportion to the 

 degree of clearing practiced locally; but it is well to recognize that published observa- 

 tions are far from being entirely reliable or irrefutable. 11 Whatever may be the diffi- 



8 Histoire de 1' Academic Royale de France," 1739, "Memoire sur le Re'tablissement 

 et la Conservation des Fore"ts." 

 Art. 220. 



10 To quote only one instance: The waters which have filtered through into the green 

 sand of the basin of the Meuse are found again at Paris at a depth of 1,798 feet (artesian 

 wells of Gr6nelle), or at 1,903 feet (wells of Passy), and this may well appear at the 

 surface of other extremely remote points. 



11 We should be led away too far if we reproduced here even a small part of the mass 

 of observations more or less precise on this subject. A great number will be found 



