^20 M. Arago on Artesian Wells 



in the same space of time, a quantity three times greater, or 

 about 39,000 cubic feet. In its medium condition, experiment 

 gives us 28,000 cubic feet per minute, which is equal to nearly 

 thirty-eight millions per day, and more than thirteen thousand 

 six hundred and fifty millions cubic feet during the course of a 

 year. This last number, we may remark in passing, is nearly 

 equal to the total quantity of rain which annually falls in this part 

 of France throughout an extent of ninety-six square mile& *. 



The most striking example which we can mention of a sulv 

 terranean sheet of water, of a varying level, is that of the Lake 

 Zirknitz, in Carniola. This lake is about six miles long by thi-ee 

 broad. Towards the middle of summer, if the season be dry, 

 its surface rapidly falls, and in a few weeks it is completely 

 dry. The openings by which the waters retire beneath the soil 

 may then be distinctly perceived, sometimes quite vertical, and 

 in other places bearing a lateral direction towards the caverns 

 which abound in the surrounding mountains. Immediately 

 after tlie retreat of the waters, all the extent of the surface which 

 they covered is put under cultivation, and at the end of a couple 

 of months, the peasants are mowing hay, ,or reaping millet and 

 rye, in the very spot where, some time before, they were fishing 

 tench and pike. Towards the end of autumn, and after the 

 rains of that season, the waters return by the same natural chan- 

 nels which had opened a passage for them at the time of their 

 departure. The description which we have just given of the in- 

 undations and retreat of the water, is the regular and common 

 occurrence ; but every extraordinary atmospheric change is apt 

 to interfere with this order. Sometimes a very heavy fall of rain 

 on the mountains with which Zirknitz is surrounded occasions 



" After heavy rains, when the fountain of Vaucluse augments very rapidly, 

 its waters have not tlieir accustomed clearness. It is therefore beyond all 

 doubt that it is the rain water which has descended the fissures, which again 

 appears at this spring. Unfortunately our information does not extend farther. 

 It has been recently alleged that the river Durance and the reservoir of 

 Vaucluse, communicate with each other under ground. But we cannot enter 

 upon the discussion of this hypothesis till we know the facts as to the compa- 

 rative levels of the bed of the river, and the bottom of the funnel whence the 

 fountain springs. Authentic documents prove, that in this well, between the 

 greatest and the least elevation of the waters, there is a difference of 66 feet 

 in the leveL Though this cuxumstance is assuredly very difficult of expla- 

 nation, yet it would be folly to suppress or disregai-d it. 



