624 
Proceedings of the Koyal Society of Edinburgh. [Sess. 
in magnetic quality was detected. The effect, therefore, is one which is 
associated only with temperature variation, and not with continued exposure 
to a constant temperature, either high or low. 
A specimen of hard steel was demagnetised by reversals, heated at an 
almost uniform rate of 20° C. per minute to 840° C., cooled at a rate of about 
12° C. per minute, and the amount of “ sensitive state ” induced measured. 
The experiment was repeated with the heating and cooling carried out at 
about a fifth of the previous rates. The variation in the temperature 
gradient was found to be without appreciable influence. 
“ Sensitive State ” and various physical constants of the material. — It 
occurred to the authors that, starting with a specimen of steel in the 
“ sensitive ” condition, the change brought about in the permeability by the 
process of demagnetising by reversals might be accompanied by changes in 
the other physical constants of the material. Experiments were accordingly 
undertaken with a view to arriving at a conclusion on this point, the con- 
stants investigated being the electrical conductivity and the moduli of 
elasticity. 
Specific Electrical Resistance. — For the resistance measurements, the 
following procedure was adopted. A helix having a length of about 150 
cms. was set up with its length east and west, and connected up in series 
by way of a reversing key with a battery and a variable resistance. The 
specimen, in the form of a wire about 140 cms. long, was carefully annealed 
from 900° C. in the furnace, placed within a glass tube having a length 
somewhat greater than that of the helix, and the whole fitted in the coil. 
The wire was connected up in series with a rheostat and a large cell of con- 
stant e.m.f. Two potential leads from a sensitive galvanometer of high re- 
sistance were connected to two points on the wire near its ends, but within 
the terminal binding screws. On passing a current through the wire, a 
deflection of the galvanometer took place, the deflection being proportional 
to the resistance of that part of the wire lying between the potential ter- 
minals. By means of the rheostat, this deflection was adjusted to a suitable 
value and kept under observation. It was found to remain perfectly steady, 
showing that practically no heating took place in the wire. The specimen 
was now demagnetised by reversals with the variable resistance and commu- 
tator. No alteration took place in the deflection of the galvanometer, 
showing that the change from the “ sensitive ” to the “ normal ” condition 
was accompanied by no change in the specific resistance of the material. 
Further experiments carried out with different specimens only served to 
confirm this result. 
Rigidity Modulus. — An examination of the elastic constants of the 
