Upon a REMARKABLE COLD. 173 







obferved fuch a thermometer in air point fenfibly below another 

 befide it, whofe ball was perfectly clear. In fliort, the whole 

 features of the one procefs, compared to thofe of the other, are 

 fo diffimilar, as to fhew evidently they depend upon principles of 

 a very different nature. 



IN the profecution of fo dark a fubjecl, we have proceeded 

 hitherto entirely in the way of analyfis, and have been careful 

 to prevent any fuppofition or conjecture from mixing with, or 

 affecting our conclufions. But having now arrived at what ap- 

 pears to be a general and leading fact, and of fo anomalovis a 

 kind, the fituation obliges us, more or lefs, to take up with fur- 

 mifes, as guides to farther experiments ; in the view of finally 

 reconciling the prefent phenomena with what we already know 

 concerning the nature and properties of heat. 



WHEN fuch a cold is produced, as that which has been con- 

 fidered at fo much length, the queftion is, What becomes of 

 the fenfible heat which thus conftantly difappears ? 



FROM the experiments mentioned in the paper of 1781, we 

 are difpofed to think, that this heat does not enter into the com- 

 pofition of the hoar-frofl which is given out by the air. 



IF, however r it fhould be found, that the air itfelf, at the 

 moment the decompofition takes place, has its capacity for re- 

 taining heat much increafed, fuch a circumftance would im- 

 mediately folve the difficulty. For, by this means, the air, 

 when parting with any given portion of hoar-froft, might ab- 

 forb, not only the whole heat, which, according to Dr BLACK'S 

 law, ferved to keep the hoar-froft diflblved, but alfo^ it might 

 abforb a portion of the fenfible heat belonging to the body or 

 furface upon which the hoar-froft fettles ; and thus occafion a 

 diminution of fenfible temperature, or the cold in queftion. 



BUT regarding the air, in thefe extreme colds, and whilft the 

 hoar-froft is diflblved in it, as a chemical mixt, it may be de- 

 manded, What reafon is there for fuppofing that the capacity of 

 the air for heat fhould increafe at the moment it gives out its 



hoar- 



