1297 
on account of the inclined position of the scale. As the distance 
from the mirror to the scale is 315 em, we find for the angular 
velocity due to one impulse 
0.21 
EED 
3315 
and (using the value 0,815 gr.em* of the moment cf inertia) 
3.3 X 10-4 X 0.815 = 2.7 X 10-4 
for the corresponding moment of momentum. 
As this is due to the change of the magnetization from — S into 
+ S, the moment of momentum corresponding to the magnetization S is 
fo. 10%. 
This!) must be equal to 1,13 107 x the magnetic moment of 
the cylinder, so that we tind for this moment 
1735><10—+ 
1 18X 107 
A rough experimental determination gave the value 
1400. 
= ILA) 5) 
6. Approximate evaluation of effect Sa. 
We observed that a deflexion of 9,2 em on the scale was caused 
by a current that did not pass through the windings of the iron 
cylinder, but only through the compensating coil and its shunt and 
the total intensity (current in tbe coil + current in the shunt) of which 
was 124,5 m.A. The resistance of the shunt was 30 &, that of the 
coil 27 2. The night before the coil had been used for the com- 
pensation II in the same position and with the same shunt, but then 
the intensity of the current had been only 29,7 m.A. If therefore 
that current had acted on the cylinder, it would have produced the 
permanent deflexion 
DNC. 
2 
The value of the remanent magnetism follows from the curve of 
fig. 2, where OD: OK = 0.14. 
Thus, the remanent magnetism is 0,14 time CA. 
The couple, with which the residual field acts on the (horizontal) 
magnetization has the same magnitude as that which gives rise to 
1) A. Emstem and W. J. pe Haas, loc. cit. p. 3. 
2) The volume of the cylinder is 150 cm*, so that this calculation gives for the 
magnetization per unit of volume 800. 
