the Magnetism of Basalt. a1 
and appeared to be much more magnetic than the Rowley 
blue basalt from the same locality. 
The Magnetic Behaviour of Bars of Basalt Heated in Air. 
Profs. Roberts-Austen and Ricker, when measuring the 
specific heat of basalt*, found that, “in the processes of 
heating and cooling, the basalt undergoes more or less 
important changes in constitution, and that frequent heatings 
au0 coolings, and the nature of the flame—whether oxidizing 
ared to affect the results very seriously.” 
The bars , being heated in air, were therefore subject to 
change of constitution, and it appeared that the greatest 
change occurred during the first heating, and diminished 
rapidly after each subsequent heating, the magnetic pro- 
perties being tolerably constant at about the sixth heating. A 
reddish, rust-like coating was formed on the surface of the 
bar, the colour increasing in intensity with the number of 
heatings, and the coating forming soonest on the Rowley | 
specimens. The above changes are accompanied by loss of 
water, more than 0°5 gram of water, or about 0°88 per cent., 
being given off in some cases. 
The “procedure by means of which the following results 
were obtained may be shortly described. Each experiment 
took about seven hours to complete, and was carried out 
mostly on Sundays, as during the week the compensation 
of the magnetic coil was liable to change from hour to hour 
owing to the wall tremors caused by the motion of machinery 
situated on the ground-floor beneath. Great care had. to be 
tuken to prevent any motion of the compensating-coil, since 
there was no means of detecting any disturbance due to this 
cause until the end of an experiment. Further, the effect to 
be measured was small, and, also, a motion of the compen- 
sating-coil was accompanied by a change in the magnetic 
field round the needle. Compensation having been effected, 
the middle of the bridge was found, this chiefly depending 
on the plug resistance of the circuit. The bar of basalt to 
be tested was then, with the platinum thermometer on the 
top, pulled carefully into its position in the furnace by means 
of thread, and the ends of the furnace stopped with cotton- 
wool to prevent air currents. Water was then passed slowly 
through the water-jacket, and corresponding thermometer 
readings and magnetometer deflexions, due to the reversal of 
a current of about 1°15 amperes in the magnetizing-coil, 
were taken as the temperature was ss raised or 
lowered. A small damping-coil, not shown in fig, 2, was 
* Roberts-Austen and Riicker, Phil, Mag. Oct. 1891, p. 353. 
HK 2 

