124 DE. W. a. DTTFFIELD, MESSRS. THOS. H. BURNHAM AND A. H. DAVIS ON 
whence 
o 
At 8 arnpere.s. 
o o 
o 
o 
At 9 amperes. 
o o 
o 
V = 175, 
126, 
190, 
186, 
178 
197 
Set A' 
Set B . 
> Mean 175 
V = 
188, 
181, 
181 
Mean 183 
E = .58, 
9.3, 
60, 
164, 
55 
22 
Set A • 
Set B . 
- Mean 
75 
E = 
94, 
101, 
97 
Mean 
97 
P = 55, 
22, 
30 
44 
Set A' 
Set B . 
> Mean 
38 
P = 
17, 
48 
Mean 
47-5 
The couples due to P, E, and V are thus approximately as 1:2:4 in the actual 
arrano'ement employed, but this is accidental. The pressures, P, at 8 and 9 amperes 
reduce to 0'91 and 1'03 dynes respectively, values in good agreement with those 
obtained by other methods. 
The Electromagnetic Effect. 
In oi'der to evaluate the couple upon the suspended arm due to the electromagnetic 
eriect of the rest of the circuit, and particularly the effect of the fixed pole at right 
angles to it, the apparatus was arranged as in fig. 17, h. The current was led to the 
centre of the copper rod from the mercury trough and from the fixed carbon by a 
vertical wire. The couple was measured for different lengths of the fixed pole and 
was found to reach a maximum at about 11 cm. Alternating current obviated any 
influence of the earth’s magnetic field. The efi'ect of altering the current strength 
with a fixed length of pole was also examined and the couple found to vary, as 
was expected, with the square of the current. These data enabled the necessary 
corrections to be made where method No. 1 was adopted {loc. cit.). 
Electrostatic Effects. 
Previous to .striking the arc the electrostatic attractions between the poles 
amounted to 0'125, 0'031, 0’008, O'OOG dynes for arcs of 1, 2, 3, 4, and 5 mm. 
respectively. But when the arc is .struck the distribution of charges within it 
entirely alters these conditions. 
In Part II. is developed a theory of the pressure which does not appear to be 
seriously affected by electric effects within the vapour. This view is supported hy 
the observation that the pressure depends upon the nature of the poles. 
(ionrexion Currents. 
Hitherto we have dealt with the total pressure upon the poles, it remains to 
consider to what extent it is caused by convexion currents of hot air and vapour 
rising from the arc. It will appear that convexion currents tend to cause the poles 
to move towards one another, and that if they could be eliminated the pressure 
would be higher. 
