330 
DR. E. H. GRIFFITHS AND MR. EZER GRIFFITHS ON THE 
The safety valve on the compressor was adjusted to blow off at the required pressure 
(the lower the temperature of the experiment the higher the blowing-off pressure) the 
engine was then again started and the temperature of the enclosure walls maintained 
at a steady temperature of about three degrees higher than that of the metal block. 
Some time had to elapse before the conditions were sufficiently settled to justify 
the commencement of an experiment. 
The first group of readings consisted of observations of the rate of rise of 
temperature of the block by radiation, the transits of the temperature being observed 
across successive equal intervals (usually 0°'04 C.) ; the time between successive 
transits being of the order of 50 seconds. 
The electrical supply was then switched on and, after allowing a little time for the 
setting up of a steady gradient, lag, &c., transits every 0°'2 C. were taken. 
When the temperature had risen two or three degrees above the surroundings, the 
electrical supply was switched off and observations of temperature and time continued. 
The temperature would rise to a maximum, then fall steadily under the influence of 
radiation ; the rate of cooling being observed in precisely the same manner as the rate 
of rise before the electrical supply. 
If a is the rate of rise or fall due to radiation for 1° C. difference in temperature 
between the block and the surroundings; 
0 O the temperature of the walls of the enclosure ; 
0, the thermometer lag for the rate of electrical supply—the magnitude of this lag 
being determined by observations of temperature and time after switching off the 
electrical supply . # 
Then assuming Newton’s law to be valid for the loss or gain by radiation (an assump¬ 
tion which was fully justified by the experimental results), we have the expression 
^ = o-(e_0 o ),.(2) 
for the rate of rise or fall under the influence of radiation alone. 
Hence, plotting dS/dt against 0, the straight line joining the two groups will cut 
the temperature axis at 0 = 0 O , which determines the temperature of the surroundings. 
For the rate of rise under the combined effect of the electrical supply and radiation, 
we have the equation 
30 
3 1 
6 being the apparent temperature. 
Plotting the observed rates of rise on the same scale as the radiation observations, 
it is obvious that the straight line thus obtained should be parallel to the line joining 
the two groups of radiation observations, since the tangent of the angle made with 
the 0 axis is equal to <r. 
* See p. 151, ‘Phil. Trans.,’ A, vol. 213, 1913, 
(nE) 2 
JR/(MS+ms) 
+ (T (0 + 0J — 0 O ), 
• • ( 3 ) 
