EXPERIMENTS and OBSERVATIONS, &c. 147 



fpongy and flaky contexture, muft expofe a vaft quantity of fur- 

 face to the action of the external air. 



THIS view of the matter appeared to me ftill more probable, 

 from having formerly been witnefs to a very quick evaporation 

 of icy films from the furface of polifhed metal at very low tem- 

 peratures ; and from other experiments, demonllrating the ac- 

 tual evaporation of ice, of which I had read an account in the 

 French Memoirs. 



HAVING always been a great admirer of Dr FLACK'S Philo- 

 fophy, in regard to heat and cold, and considering the prefent 

 phenomenon as an immediate confequence of his law of eva- 

 poration, I refolved to attend to it further, in all its circum- 

 ftances, from a defire of extending a little the boundaries of 

 that induction which, in his hands, has been fo fertile in gene- 

 ral principles of fo much importance. 



IN the two papers above quoted may be feen by what fteps, 

 as opportunities offered, I followed out this defign, and how 

 very early I was perplexed by facts which feemed ftrongly to 

 evince, that no evaporation whatever was going on when the 

 excefs of cold, at the furface of the fnow, was perceivable. 

 On the contrary, the further the experiments were puflied, the 

 ftronger were the prefumptions, that the air, fo far from ab- 

 forbing from the fnow, or wafting it by evaporation, was all 

 the while depofiting hoar-froft profufely upon its furface, as 

 well as upon all other bodies with which it had a free commu- 

 nication. 



THE following extract from the paper of the year 1780, will 

 fet this in a ftrong light, and will mow, how far my firft im- 

 preffions were obliged to yield to the authority of facts. 



' ON Sunday night, January 23. 1780, feveral things were 

 ' laid out at the Obfervatory, fuch as flieets of brown paper, 

 ' pieces of boards, plates of metal, glafles of feveral kinds, 

 ; &c. which all began to attract hoar-froft, feemingly as foon 



as they had time to cool down to the temperature of the air. 



T 2. " The 



