J34 Tkfsry of thi Cold Winds wkuh iJpic/t\>Ni the Ejyth. 



He thinlts, therefore, that evaporation would not fuffice to explain tlie refrigeration of 

 fevea or eight degrees below temperate, fuch as is obfcrvcd in the caves of Lugan, but that 

 it would fuffice to explain a cold cf five or fix degrees below this term, fuch as that at 

 Cefi, Ifchia, and Mont Teftacco. In faifl, he fuppofcs a great fubterraneous rcfervoir of 

 air to exill fufficiently near the furfacc of the earth, that the cold of the winter may caufe 

 it to dcfcend three degrees below temperate, or the tenth degree, and that the heat of 

 fummer may caufe it to rife as much above that term. 



When the cold fliall have penetrated this rcfervoir to its maximum, the temperature will 

 be feven degrees. 



Afterwards, when the heat of fpring begins to dilate it, its temperature will rife, fuppofe 

 to eight. It will begin to iflue out ; and the evaporation, diminifhing its heat three degrees, 

 will reduce it to five ; and this will be its term of the preatcft cold. In proportion as the 

 heats of fummer (liall penetrate the refervoir, the heat of the air which iflues out will in- 

 creafe alfo : it will not however rife above temperate, becaufe the grcateft heat of the rc- 

 fervoir will be 13, and evaporation will reduce it to 10. 



The comparifon of his experiments with each other, as well as with thofe of the Abb« 

 Kollet, proves, in facY, that the heat of thefe cool winds increafes as the feafon advances. 

 For the Abbe NoUet found the oaves of Mont Teftacco at 9,5 on the 9th of September, 

 whereas M. De SaufTure obferved them 115,3 on the iftof July. In the fame manner 

 he found the caves of Lugan at /\\ on the til of Auguft, and at 23- a month earlier, namely» 

 on the 29th of June. 



When the air has obtained its highell degree of heat, it muft remain for a certain time in 

 a ftate of ftagnation; after which the refervoir begins to cool, and abforb the external air. 

 In this fituation the coolnefs of the autumn, and the fvofts of winter, are fufficient to pre- 

 ferve the cold ftate of the caves. 



The author remarks, that from the original fuppofitionof the medium temperature being 

 at 10 degrees, and the cold produced by evaporation 3 degrees, it is impofliblc to explain a 

 degree of cold which in fummer is lower than five degrees. For if the refervoir be fup- 

 pofed nearer the furface, fuch for ex.imple as that the cold of the winter would caufe it to 

 defcend to 5, it is true that the air with the additional cooling by evaporation would iflue out 

 in the fpring at the temperature of two degrees, but it will likewife follow, that the refer- 

 voir would rife in its temperature proportionably higher by the heat of fummer. 



If, therefore, it be required to explain a greater degree of coolnefs th.m 4 or 5 degrees,, 

 fuch as that of Lugan and Hergifweil, and it be not fuppofed that evaporation in thefe cir- 

 cumftances can produce more refrigeration than three degrees, it will be necclTary to fup- 

 pofe the mean temperature of the refervoir to be lower than 10 degrees; a fuppofition, as 

 the author obferves, by no means forced, at leaft for the vicinity of the Alps } where alone 

 obfervations have been made of caves pofieffing fo low a temperature. 



M. De Sauflure anticipates the objedion againft his theory, that if the air in thefe ca- 

 verns be already faturated with humidity, it cannot produce evaporation, nor confequently 

 cold. But he remarks, that caverns are not all humid; that the caverns here fuppofed mufi 

 be of vaft extent, in order that the dilatation caufed by a few degrees fhall afford confuler- 

 ablc currents of air through the whole fummer; and confequently, that a great quantity of 

 cold dry air muft entei iin with the winter, which will be very deficcative when expanded 



by 



