326 On radiation, absorption, ¥c. 
temperature of equilibrium, between the heat absorbed, and that 
which is radiated. If the vessel be now turned, so that the various- 
ly coated surfaces are towards the source of heat, the liquid in those 
coated with the best absorbents, will immediately begin to rise in 
the tubes, and that in those coated with the worst absorbents, to fall. 
That the two lateral compartments are exposed to a greater cooling 
action than the others, may be an objection to this apparatus ; but it 
is easily obviated, and with it the communication of heat from one 
compartment to another, by terminating the box at each end bya 
small compartment, and separating each of the other compartments 
by a similar space; in fact, convenience alone was the reason for 
uniting these air thermometers in one vessel. 
nother form of apparatus, which is more Fig 4. 
simple, I have found convenient ; but it occu- 
pies more time than that last described, in ob- 
taining the same results. A prism of any con- 
venient number of sides, is made into an air 
thermometer, in the manner described in speak- 
ing of the last apparatus ; and the sides are va- 
riously coated ; it fits loosely into a prism of 
the same form, but wanting one side; in the 
figure, a b ce, represents the eveloping sur- 
face, and m no p, the air thermometer. To 
show the different absorbing powers of the different substances, the 
vessels described are placed as in the figure, before another, A, contain- 
ing hot water, hot sand, or any other convenient source of heat. Sup- 
pose the side cf the air thermometer which is the worst absorbent 
of heat, to be exposed to the source of heat, the air within is expan- 
ded, and the position of the liquid in the tube is marked by an index ; 
a better absorbent is exposed, and the liquid rises higher ; a worse, 
and it falls below its original level ; the experiment can thus be va- 
ried at pleasure. The outer sheath, or covering prism, serves to ren 
der the surface, not exposed to the source of heat, uniform in its 
radiating powers, and to protect those sides which are not intended 
to be exposed to the source of heat, from the radiation of the ves- 
sel, A, which, otherwise, would affect them sensibly. If the ar 
thermometer were a rectangular prism, of course the objection just 
stated would not apply ; but the sheath would still be necessary t0 
= the radiation from the surfaces not exposed to the source of 
eat. 
