Inquiries concerning the Mode, etc. 275 



When a solid body is plunged in a liquid which is 

 hotter than the body, the particles of the liquid in con- 

 tact with the body, being condensed by the cooling they 

 undergo, descend, in consequence of the increase of 

 their specific gravity, and fall to the bottom of the 

 liquid ; and the strata situated above the level of the 

 cold body are not cooled by it immediately. 



It is true that the viscosity of liquids, even of those 

 which possess the highest known degree of fluidity, is 

 still much too great to allow one of their particles indi- 

 vidually being moved out of its place by any change of 

 specific gravity occasioned by heat or cold ; yet this does 

 not prevent currents from being formed, in the manner 

 above described, by small masses of the liquid composed 

 of a great number of such particles. 



The existence of currents in the ordinary cases of the 

 heating and cooling of liquids cannot any longer be called 

 in question ; but philosophers are not yet agreed with 

 respect to the extent of the effects produced by those 

 currents. 



In treating of abstruse subjects, it is indispensably 

 necessary to fix with precision the exact meaning of the 

 words we employ. The distinction established between 

 conductors and non-conductors of heat is too vague not to 

 stand in need of explanation. An example will show 

 the ambiguity of these expressions. 



If two equal cubes of any solid matter, copper, for 

 instance, of two inches in diameter, the one at the tem- 

 perature of 60, the other at 100, be placed one above 

 the other,- the cold cube will be heated by the hot one, 

 and this last will be cooled. 



If the cold cube be placed upon a table and its upper 

 surface covered by a large plate of metal, of silver, for 



