258 
MESSRS. R. T. GLAZEBROOK AND E. B. SARGANT ON THE 
temperature of 12°, are given below; the temperature is that recorded by a thermo¬ 
meter laid on, or on some occasions inside, the box. Each number is the mean of 
three or four measurements taken at the same time. 
Plug out. 
Value. 
Observer. 
f 9-9903 
R. T. G. 
10 
9-9902 
R. T. G. 
L 9-9890 
E. B. S. 
r 9-9887 
R, T. G. 
10' < 
9-9907 
R. T. G. 
1 
L 9-9908 
E. B. S. 
1+2+2+5 
9-9870 
E. B. S. 
20 | 
'19-9772 
R. T. G. 
. 19-9775 
E. B. S. 
10 + 10' 
19-9769 
R. T. G. 
1-10 
19-9756 
R. T. G. 
For the Warden and Muirhead box the differences between two sets of experi¬ 
ments were quantities of the same order as here. 
The two boxes were then placed at opposite ends of the bridge-wire, the other two 
arms of the Wheatstone’s bridge being the pair of 30-ohm coils used in the previous 
part of the experiments. 
The 50-unit plug was taken out of one box, and out of the other all the plugs from 
1-20. The difference between these two nominal 50 units was thus obtained. Then 
plugs 1-20 were taken out of the first box, and the 50 units out of the second, and 
another difference obtained. In this manner the values of the coils 50, 100, 100', and 
200 were obtained. 
After this the differences between the two boxes became too large to be measured 
in terms of the resistance of the bridge-wire, and recourse was had to the method 
employed in Part I., by which one coil in a box was compared with-a combination of 
coils in the same box. 
Two coils, each of about 5 ohms, were connected with the ends of the bridge-wire, 
while the coils to be compared, P and Q, formed the other arms of the bridge. 
Let 5 + «, 5 + /3 be the resistances of the two 5-ohm coils, and let x, y, x , y' have 
the usual meanings. 
Then 
P_5 + a + x 
Q~5 + /3 + y 
Intercharge the 5-ohm coils 
P 5+/3+P 
