332 
DR. E. H. GRIFFITHS AND MR. EZER GRIFFITHS ON THE 
IV. Cooling Curve. 
(Time measured from instant of switching off, and temperature difference from the 
last transit.) 
Time. 
Temperature 
difference. 
Time. 
Temperature 
difference. 
secs. 
secs. 
246 
+ 0-120 
698 
0 
331 
o-ioo 
786 
-0-020 
396 
0-080 
854 
-0-040 
481 
0-060 
942 
-0-060 
545 
0-040 
1011 
-0-080 
633 
0-020 
If the results of IV. are plotted on sectional paper and the line prolonged to cut the 
temperature axis for t — 0, we obtain the “rise above ” after switching off, which 
quantity is denoted by in the equations above. In this particular case = O' 18 5° C. 
Referring to fig. 3, the observations at “ A ” are the rates of rise by radiation in 
the above experiment, when the temperature of the block was below that of the 
surroundings; those at “B” correspond to the rate of cooling under radiation after 
the block has been heated up above the surroundings by the electrical supply. 
In some experiments the electrical supply was switched off; a group of “ radiation ” 
observations taken ; then the electrical supply continued and the experiment completed 
in the usual way. 
The intermediate group of “ radiation ” points thus obtained always fell on the 
straight line joining the extreme groups and justified the application of Newton’s law 
for the small temperature differences employed. 
The observations distributed about the line “ CD ” correspond to the period of the 
electrical supply. It will be observed that, as required by theory, “ CD ” is parallel 
within the limits of experimental error to the line “ AB.” 
If desired, the radiation group at “ A ” could be dispensed with and the line drawn 
through the mean of group “B ’ parallel to “CD.” Occasionally this plan was 
adopted when the observations “ A ” were very irregular on account of the conditions 
not being sufficiently steady when the observations were commenced. 
The point “ N ” on AB ” corresponds to the temperature at which there is no loss 
or gain by radiation and “ NQ ” to d0 N / dt of equation 4. 
In the case of several metals, experiments were performed near 0° C. by this 
method and also by the “ intersection ” method described in the previous paper, with 
entire accordance between the results. 
(8) Experimental Results. 
The results obtained in this and the previous investigation are summarised below. 
Where more than one experiment has been performed at any temperature, the mean 
