19 Radiation of Cold. 
ing that there are rays of heat, and that they are subject te 
the same law of reflexion as those of light. Now if we re- 
move the hot body, and place in its stead a mass of ice, (the 
thermometer beitig supposed to have regained the tempera- 
ture of the room,) rays of cold will apparently proceed from 
it, reach the thermometer by the same route as before, and 
cause it to descend. Why should we infer such a principle 
as heat in the one case, more than such a principle as cold in 
the other? ve we not here the same evidence of the ex- 
istence of rays of cold, as we had before of rays of heat ? 
Of the various hedbas which have been given of this 
phenomenon, i in accordance with the supposition that heat is 
the only positive principle, the greater part appear to be alto- 
gether unsatisfactory. The only explanation which appears 
to me to throw any light on the subject, is that which ascribes 
the depression of temperat ure to the circumstance, that @ 
es of heat is intercepted by the cold body which would 
herwise be conveyed to the thermometer. Ef we narrowly 
Soinider the circumstances, we shall perceive that all the heat 
which is conveyed to the thermometer by the mirrors, comes 
through that point which is now oceupied by the cold body, 
and consequently is intercepted by that. Thus, 
Let A B be two parallel concave reflectors. Let a ther~ 
mometer be placed in the focus 'T, and first, let the focus H 
remain unoccupied. Now of all the Bios of heat, proceed- 
ing from every direction and falling on the mirror A, none 
will be conveyed to the thermometer buf such as come to A 
parallel to each other. But if we follow back these rays, 
we shall find that they are the same that had previously passed 
gh the focus H. Now let the cold body be placed in 
this focus, and it is evident that the whole body of roa 
rays of heat, which Agia through this point, will be cut 
while the cold body itself does not radiate an equal cxarouist 
of heat, and hcttbire oceasions a loss of heat to the thermo- 
meter. That a depression of the thermometer is occasioned 
