ON THE SOURCES AND EFFECTS OF HEAT. 487 



one sixth part of its weight of charcoal, twelve pounds of ice may be 

 melted.* 



The manner in which heat, when excited or extricated by any of these 

 means, passes from one body to another, requires to be very particularly 

 examined. We shall find that this communication happens in one or both 

 of two ways, by contact, or by radiation ; and that it may also differ both 

 with respect to the quantity of heat concerned, and to the time occupied 

 by the process. Whatever heat may be, we may safely conclude that in 

 substances of the same kind, at the same temperature or apparent degree 

 of warmth or coldness, its quantity must be proportional to the mass or 

 weight ; for instance, that a quart of the water of a given cistern contains 

 twice as much heat as a pint ; and where this is true of the different parts 

 of any substance, they must remain in equilibrium with respect to heat. 

 But if two equal portions of the same substance, containing different 

 quantities of heat, be in contact, they will affect each other in such a 

 manner as to have their temperatures equalised, and the more rapidly as 

 the contact is more perfect. Thus, if two portions of a fluid at different 

 temperatures be mixed together, they will acquire immediately an inter- 

 mediate temperature ; and when two solids are in contact, the quantity of 

 heat, communicated by the hotter to the colder in a given time, is nearly 

 proportional to the difference of the temperatures. Hence it would follow, 

 that they could never become precisely of the same temperature in any 

 finite time ; but in fact the difference of temperature is rendered, in a mode- 

 rate time, too small to be perceptible. The nature of the substances 

 concerned has also a material effect on the velocity with which heat is 

 communicated through their internal parts ; metallic bodies in general 

 conduct it the most readily, earthy and vitreous bodies the least ; but the 

 various metals possess this power in different degrees ; silver and copper 

 conduct heat more rapidly than iron, and platina transmits it but very 

 slowly. Professor Pictet supposes that heat ascends within solid bodies 

 more readily than it descends ; but the effect of the air remaining in the 

 imperfect vacuum of the air pump may be sufficient to explain his experi- 

 ments ; the difference of temperature producing an ascending current 

 in the neighbourhood of the heated body, by means of which the cold ail- 

 continually approaches its lower parts, and carries the heat upwards : and 

 it has been found that the rarefaction of air does not by any means 

 diminish its power of conducting heat, in proportion to the diminution of 

 its density. 



Count Rumford's experiments t have shown that all fluids are very 

 imperfect conductors of heat by immediate contact, although it is scarcely 

 credible that they can be absolutely nonconductors ; but heat is usually 

 communicated between different portions of the same fluid, almost entirely 



* On combustion, consult Hooke, Micographia, p. 103. Lavoisier and Laplace, Hist, 

 et Mem. 1780, p. 355, H. 3. Rumford, Nich. Jour, xxxii. 105 ; xxxiv. 319 ; xxxv. 

 95. Davy, Ph. Tr. 1817. Sym, Annals of Ph. viii. 321. Davies, ibid. (2nd series), 

 x. 447. Dobereiner, Schweigger's Jahrbuch, iv. 91 ; viii. 321. 



t Ph. Tr. 1786, p. 273 ; 1792, p. 48. Essays, Lond. 1796. See also Dalton, 

 Manch. Mem. v. 373. Thomson, Nich. Jour. iv. 529 ; 8vo, i. 81. Murray, ibid, 

 i. 165, 242. Trail, ibid. xii. 133. Despretz, Comptes Rendus, vii. 933. 



