118 On the Powers of Steam, fcfc. 



phere is loaded with clouds or moisture, it is consi- 

 derably diminished. This shows evidently, that al- 

 though the air is not very cold, yet if it is in a fit state 

 to promote the evaporation of the moisture which 

 transudes the vessel, it will succeed in producing ice. 

 Supposing that these cloudy nights had always been 

 windy likewise, still it is scarcely to be supposed, 

 that the water, in such small vessels, placed at least 

 a foot below the surface of the earth, could suffer such 

 agitation as to prevent freezing ; more especially when 

 we consider, that some degree of motion in a fluid is 

 found, by experiment, to be rather favourable to its 

 freezing*. In large bodies of water, such as rivers, 

 lakes, &c, the reverse must necessarily take place, 

 because all agitation tends to mix the coldest parts of 

 the water with those which are less cold ; so that 

 unless the winter is long enough to reduce the whole 

 body to 32 degrees, no freezing can take place whilst 

 there is any considerable agitation. 



* Dr. Black has found, that water which has been previously 

 boiled freezes sooner than unboiled water; but if the unboiled 

 water is stirred about, it will freeze as soon, or nearly as soon, as 

 the boiled water. Muschenbrock has also observed, that water 

 would fall below 32 without freezing, provided it was kept per- 

 fectly motionless ; but if it was agitated, it was immediately con- 

 verted into ice. Fahrenheit inclosed water, purged of its air, in 

 glass globes hermetically sealed, and observed, that if they were 

 kept motionless, the water became colder than the freezing point, 

 without losing its fluidity ; but as soon as the globes were shook 

 or stirred, a sudden congelation ensued. 



