478. IN the previous chapter it was found that the classical system of mechanics, when applied to the radiation problem, led to a solution which proved to be in violent disagreement with experience. We shall begin the present chapter by explaining a different line of attack on this problem. This will be based only on general thermodynamical principles, and it will be noticed that it does not require any assumption to be made as to the existence, or non-existence, of an ether.

Consider an enclosure of any shape, of which the walls are impervious to energy of all kinds, and therefore in particular to radiation. Let it contain a certain amount of heated matter which will of course fill the enclosure with radiant energy, and let a steady state finally be reached in which the matter is at a temperature T. Since there is no loss of energy in this state, each piece of matter inside the enclosure retains its temperature indefinitely, and therefore the amount of energy it gains by absorption of radiation must exactly balance the amount it loses by emission of radiation. Considering two pieces of matter, A and B, it is readily seen that the stream of energy which flows from A to B must be exactly equal to that which flows from B to A. Hence we arrive at the conception of a stream of energy appropriate to matter of temperature T, this depending only on Tand not on other quantities involved in the structure of matter. It follows that the density of radiant energy inside the enclosure will be a function of T only.