62 HISTORY of the SOCIETY. 



him every day converfe with no lefs clearnefs and precision, 

 than animation and force. From whatever caufes the want of 

 perfpicuity in his writings proceeded, perplexity of thought was 

 not among the number ; and the confufion of his ideas can nei- 

 ther be urged as an apology for himfelf, nor as a confolation to 

 his readers. 



Another paper from his pen, a Theory of Rain, appeared alfo 

 in the firft volume of the Edinburgh Tranf actions. He had long 

 ftudied meteorology with great attention ; and this communi- 

 cation contains one of the few fpeculations in that branch of 

 knowledge entitled to the name of theory. 



Dr Hutton begins with fuppofing that the quantity of hu- 

 midity, which air is capable of difTolving, increafes with its tem- 

 perature. Now, this increafe muft either be in the fame ratio 

 with the increafe of heat, in a lefs ratio, or in a greater : in 

 other words, for equal increments of heat, the increments of 

 humidity muft either conftitute a feries of which all the terms 

 are equal to one another, or a feries in which the terms conti- 

 nually decreafe, or one in which they continually increafe *. If 

 either of the two firft laws was that which took place in nature, 

 a mixture of two portions of air, though each contained as 

 much humidity as it was capable of duTolving, would never 

 produce a condenfation of that humidity. According to the 



firft 



* To fpeak ftri£Uy, the law which connects the increments of humidity in the 

 ak with the increments of temperature, is not confined to any one of the three 

 fuppofitions here made, but may involve them all. The humidity diflblved may 

 be proportional to fome fund, 'ion of the heat, that varies in feme places falter, and 

 in others flower, than in the Ample ratio of the heat itfelf. Neverthelefs, for that 

 extent to which obfervation reaches, the reafoning of Dr Hutton is quite fufficient 

 to prove that it varies falter ; or, in other words, that if a curve be fuppofed, of 

 which the abfciifce reprefent the temperature, and the ordinates the humidity, this 

 curve, though it may in the courfe of its indefinite extent be in fome places con- 

 cave and in others convex toward the axis, is wholly convex in all that part with 

 which our obfervations are concerned. 



