898 
the potential difference between the electrodes ¢,,, e,, of the constantin 
wire, are applied outside the calorimeter vessel near the sealing 
places. The resistance between these junctions and the ends of the 
constantin wire is inappreciable compared with the resistance of 
the latter (140 &). For the gold wire, which serves as a thermo- 
meter, potential as well as current wires (within the calorimeter 
vessel 0.1 mm. platinum, insulated with silk) lead directly to the 
electrodes. 
The stopcocks 4, and 4, have wide borings, 4, leads to a GaAupE 
mercury pump, which serves for evacuating the charcoal and for a 
preliminary evacuation of the calorimeter vessel, and to a Mc.-Leod 
gauge, R is a reservoir which is filled with hydrogen; with the 
aid of the pipette p a quantity of this hydrogen is admitted into 
the calorimeter vessel when it is required to bring about heat 
conduction between the block and the bath. 
The charcoal tube A, made of glazed opaque quartz, was filled 
with cocoa-nut charcoal, and evacuated for a sufficient time at 
600° C. before each series of measurements. During the measurements 
it is cooled with liquid hydrogen. 
To diminish the exchange of heat between the metal block and 
the bath through the conducting wires, during the measurements 
the bath that surrounds the calorimeter was brought as nearly as 
possible to the temperature at which the measurement was to be 
made. The time-rate of the temperature before the heating and 
after the heating, at least after a few tens of seconds, was as a 
rule very small, if not inappreciable. For illustration in Fig. 3 and 4 
+ I 
SIEN A re ee 9+ Tree e eee eee ee 
| 
| 7 
Site Jr 
L 4 
! 
na 33+ 
| 
35+ 
ene | 
5 = + + + 4 BT lessees sees ; J , 
) 35 ) 35 () IS TINA ces 2 Br) 8) hr] $0 40 50X38 
Fig. 3. Fig. 4. 
the time-curve of the galvanometer, which indicates the temperature, 
during two measurements is represented, viz. for a measurement of 
the specifie heat of lead (Fig. 3) and for a measurement of the 
specific heat of copper (Fig. 4), respectively. The increase of tempe- 
rature was in both cases about 1 degree. In the measurement with 
