496 INFLUENCE OF AGRICULTURE ON CLIMATE. 



been used as moving pov crs, have very sensibly diminished. In 

 other places, the rivers are said to have shrunk visibly ; and in 

 others, springs that were formerly abundant, have almost dried up. 

 Observations to this effect appear to have been principally made in 

 valleys, surmounted by mountains ; and it is generally asserted, tliat 

 the falHng off in the springs and streams, had followed close upon 

 the period at which the woods, scattered over the surface of the 

 country, were cleared away without any kind of reserve. 



These statements, which may be assumed as facts, seen to indi- 

 cate that w^here the woods have been felled, it rains less than it did 

 formerly ; this, indeed, is the general opinion entertained on the 

 subject ; and were it admitted, without further examination, the 

 natural inference from it would be, that the extension of agriculture 

 diminishes the annual quantity of rain which falls in a country. But 

 at the same time that the facts as stated have been observed, it has 

 further been noticed that since the clearing of the surface from for- 

 ests, the torrents and rivers which seemed to have lost in amount 

 of regular supply of water, had hecome subject to sudden and extra- 

 ordinary risings which had proved the cause of numerous and grave 

 calamities. In the same way, springs that are generally all but dry, 

 have been seen to burst forth again abundantly after violent storms. 

 These latter observations, as may readily be imagined, are of a kind 

 that should lead us not lightly to embrace the vulgar opinion, which 

 maintains that the cutting down of the woods has had the effect of 

 lessening the mean antmal quantity of rain : it is not only not impos- 

 sible that this quantity has not varied, but it may even happen that 

 the mass of water which passes over the bed of a stream, supposed 

 shrunken, is actually the sanje as ever it was; the only difference 

 may be, that now the flow is much less regular than it used U) be : 

 in former tinjes the bed was always and more moderately full ; at 

 })resent it is excessively full at intervals only. It is very possible, 

 tlierefore, that here as elsewhere, occasionally, the a})pearance of 

 the fact has been taken for the reality. It were very important to 

 discover some natural index to a solution of the question at issue : 

 whether or not the destruction of the forests that once covered the 

 face of a district of country, had had the efTecl of lessening the mean 

 annual fall of rain 1 



Tlio lakes which are met with in plains, and at different levels in 

 mountain ranges, seem to me peculiarly well calculated to throw 

 light on this subject. Lakes may, in fact, be received as natural 

 gauges of the ruiming waters of a country. If the mass of tin? water 

 contained in the lakes undergo change in otie direction or another, it 

 is obvious that this clrange, and the direction in which it has occur- 

 red, will be proclaimed by the state or mean level of the like or 

 lakes, which will ditTer for the same reason that it does at dilTtrent 

 seasons of the year, viz. as drought or rain prevails. The uican 

 level of the lake or lakes of a district will, therefore, fall, il' the 

 quantity of water which flows through that district iliminishes ; the 

 level, on the contrary, will rise, if its streams increase ; and it will 

 remain stationarv if the afllux and efflux of the lake contir.ue un 



