478 Scientific Proceedings, Royal Dublin Society. 



transference of energy at a certain rate away from the lampblack. 

 If we could calculate the whole energy of either of these radiations? 

 it would tell us the whole energy per cubic centimeter of the ether 

 considered as a medium transmitting radiations in a particular di- 

 rection, and six times this amount would evidently be the total 

 heat per cubic centimeter of the ether at this temperature. This 

 very calculation has been performed approximately for the solar 

 radiations, where we can approximately measure the total radiation 

 in a particular _ direction ; but as we only know very roughly, in- 

 deed, what is the temperature of the surface of the sun, we can 

 only calculate very roughly what is the total heat of the ether per 

 cubic centimeter at that temperature. As in addition we don't 

 know the laws of cooling for any sufficiently large range to apply 

 it safely to the sun, we could only very roughly, indeed, calculate 

 the total heat of the ether per cubic centimeter at temperatures 

 with which we are in the habit of dealing. We can, however, from 

 experiments on the rate of cooling of lampblack, calculate pretty 

 accurately the difference between the energy per cubic centimeter 

 of the ether at slightly different temperatures, and this tells us the 



do 

 specific heat of the ether at this temperature. If — be the quantity 



of heat per second lost by each square centimeter of surface of 

 lampblack when its temperature is 1° 0. hotter than its surroundings 

 at the temperature of 0° 0., then this energy is distributed over V 

 cubic centimeters of ether where V is the velocity of light in 



centimeters, and consequently — = — --- is the equation by which 



to calculate c, the specific heat of the ether. The 6 comes in owing 



to this being the energy required to raise the temperature of the 



ether by radiations going in only one direction. From several 



sources which, however, only approximately agree with one another, 



da 

 I gather that — = "0026. The great difficulty in determining this 



G/t 



is to distinguish between the heat carried off by the gas present, 

 and that lost by radiation alone. As F = 3 x 10^°, we see that 

 c = 5*2 X 10"'^ Of course, as was to be anticipated, this is a very 

 small quantity, and is very small even compared with the specific 

 heat per cubic centimeter of the gas in a good vacuum. The specific 

 heat per cubic centimeter of an air vacuum of one millionth of an 



