FRESrCIPLES OF HEATI^-Q HOT-HOUSES. 159 



with heat — by both modes, increases in proportion as the 

 heated body is of greater temperature above the surrounding 

 medium. 



The cooling of a heated body, under ordinary circumstances, 

 is evidently the combined effects of radiation and conduction ; 

 the conductive power of the air is, evidently, owing to the ex- 

 treme mobility of its particles, for otherwise it is one of the 

 worst conductors with which we are yet acquainted, so that 

 when confined in such a manner as to prevent its freedom of 

 motion, it becomes useful as a non-conductor. 



The proportion which radiation and conduction bear to each 

 other has, in general, been very erroneously estimated. Count 

 Rumford considered the united effect, compared with radiation 

 alone, was as five to three, and Franklin supposed it to be as 

 five to two. 



No such general law, however, can be deduced, for the relative 

 proportions vary with the temperature, and with the peculiar 

 substance, or surface, of the heated body; for, while the cooling 

 effects of the air, by conduction, is the same on all substances, 

 and in all states of the surface of those substances, radiation 

 varies very materially, according to the na^^ure of the surface. 



The influence of the air, by its power of conduction, varies 

 also with its elasticity. The greater its elastic force, the greater 

 also is its power of cooling, according to the following law : — 

 When the elasticity of the air varies in a geometrical progres- 

 sion whose ratio is 2, its cooling power also changes in a geo- 

 metrical progression whose ratio is 1.366. 



The same law holds with all gases, as well as with atmos- 

 pheric air ; but the ratio of the progression varies with each gas. 



To show the relative velocities of cooling at different temper- 

 atures, the following table, constructed from the experiments of 

 Petit and Dulong, is given. The first column shows the excess 

 of temperature of the heated body above the surrounding air ; 

 the second column shows the rate of cooling of a thermometer 

 with a plain bulb, and the third column giv^es the rate of cooling 

 when the bulb was covered with silver leaf. The fourth column 

 shows the amount due to the cooling of the air alone ; and by 



deducimg this from the second and third columns respectively 



14=^ 



