1913-14.] The Hall and Transverse Thermomagnetic Effects. 213 
Now (N — M) is the thermoelectric power of copper with respect to 
constantan, and is known from the results of a special experiment. The 
value of SO can therefore be calculated from the observations <5E and Se c 
Again, by (1) and (2), 
E_M 
e ~N’ 
so that (3) and (4) may be written 
E + SE = V - M(0 + SO) 
(e + $e)x- = Vx?-M(0 + S0), 
e e 
from which 
and hence 
y _e8E-E8« 
e - E 
The transverse potential difference V can thus be determined from the 
observations E, e, <5E, and $e. 
Hall Effect. 
The tubes and jackets were supplied with steam or cold water in order 
to bring the plate to the required temperature. A current of 10 amperes 
was passed through the plate, and the temperatures at E, B, and D were 
measured ; observations being taken also with the current reversed, in order 
to correct for any direct action of the current on the E.M.F.’s at the 
junctions. 
The copper wires at A and C were then connected to one galvanometer 
circuit and the constantan wires at A and C to the other, and readings 
were taken with magnetic field zero and also with known magnetic fields 
in each direction. From these readings the transverse E.M.F. due to the 
magnetic field could be calculated. 
The potential gradient along the axis of the plate was determined by 
connecting the copper wires at E and D to one galvanometer and the 
constantan wires at E and D to the other, and taking readings with both 
directions of the current. 
The value of R was then calculated in accordance with the definition 
given. 
No correction was made for the influence of the Transverse Galvano- 
magnetic Temperature Effect, as, in the case of the metals tested, this 
effect is very small. 
