HEAT OR CALORIC. 77 



temperatures, an incessant circulation is supported, so long as any 

 one portion of the liquid is cooler than another ; or in other words, 

 till the water boils ; previously to which, every particle must have 

 combined with as much caloric, as it can receive, without being con- 

 verted into steam." 



" The manner in which caloric is distributed throughout liquids by 

 circulation, as above described, is illustrated advantageously by an ex- 

 periment contrived by Rumford, who first gave to the process, the at- 

 tention which it deserves." 



"Into a glass nearly full of water, as represented by the foregoing 

 figure, some small pieces of amber are introduced, which are in spe- 

 cific gravity so nearly equal to water, as to be little influenced by grav- 

 itation." 



" The lowermost part of the vessel being subjected to heat, while 

 thus prepared, the pieces of amber are seen rising vertically in its 

 axis, and after they reach the surface of the liquid, moving towards 

 the sides, where the vessel is colder from the influence of the exter- 

 nal air. Having reached the sides of the vessel, they sink to the 

 bottom, whence they are again made to rise as before. While one 

 set of the fragments of amber, is at the bottom of the liquid, some 

 are at the top, and others at intermediate situations ; thus demonstra- 

 ting the movements, by which an equalization of temperature is ac- 

 complished in liquids." 



" When the boiling point is almost attained, the particles being 

 nearly of one temperature, the circulation is retarded. Under these 

 circumstances, the portions of the liquid which are in contact with 

 the heated surface of the boiler, are converted into steam, before 

 they can be succeeded by others ; but the steam thus produced, can- 

 not rise far before it is condensed. Hence the vibration and singing, 

 which is at this time observed." 



8. Provide a glass tube twelve or fifteen inches long and from 

 two to two and a half wide, closed at one end, and that end thin, so 

 as to bear heat ; nearly fill the tube with alcohol, and then with a 

 dropping tube, convey to the bottom some alcohol, colored by turmeric 

 or cochineal and rendered a little heavier by water ; if dexterously 

 done, there will be a well defined line of separation ; then apply heat 

 at the bottom, and the color will be rapidly diffused. 



Now repeat the experiment, only place the colored alcohol* on the 

 surface ; the color on the top will be scarcely disturbed till the fluid 

 begins to boil. 



* Jt is hardly necessary to say, that no water should be added to it. 



