470 ON IMPRESSIONS OF COLD 
rent of three miles an hour would accelerate tenfold the trans- 
fer of heat. 
It isof essential importance, however, to remark, that in all 
cases where the medium of communication is a liquid, the na- 
ture of the hot surface, whether vitreous or metallic, rough or 
polished, has no influence whatever m modifying the rate 
with which the heat is drawn off and dispersed. Thus, 
a hollow tin ball filled with hot water, will lose its heat just 
as fast when immersed naked in cold water, as if it were cover- 
ed with linen, or paper, or a coat of any sort of pigment. 
3. Heat is transferred through a gaseous medium by a more 
complex process. It is partly conducted, as in the case of li- 
quids, by successive communication through the stagnant mass, 
joined to the more copious effect of the buoyant streaming of 
the heated portions of the fluid. But another auxiliary prin- 
ciple, depending merely on the nature of the heated surface, 
now comes into action. Let two equal hollow balls of bright 
thin silver, the one naked, and the other covered with a fold of 
cambric or paper applied closely to its surface, be filled with 
warm water, and sitspended in a close room. The former 
will be found to lose ii: parts of heat, in the same time that 
the latter sheds 20 parts. Of the heat thus spent, 10 parts, 
from each of the balls, are communicated in the ordinary 
way, by the slow recession of the particles of air, as they 
come to be successively heated by their proximity. The re- 
maining portions of heat, consisting of 1 part from the bright 
metallic surface, and 10 parts from the cambric, are pro- 
pagated through the aérial medium, by some peculiar pro- 
cess. In like manner, two equal hollow balls of glass, the one 
gilt with gold or silver leaf, being filled with warm water, and 
suspended in a room, the naked one will discharge 13 parts 
of heat, while the gilt one will lose only 7 parts; 6 parts be- 
ing spent by each, in the same way as if they had been im- 
mersed 
