625 
1907-8.] Sensitive State in Magnetic Materials. 
material in the two conditions was carried out in the following manner. 
A wire about 150 cms. in length, which had been annealed from 900° C. r 
was attached to a rigid wall-bracket, and carried at its lower end a cylindri- 
cal lead vibrator. The wire was surrounded by a long helix, so that it might 
be put through the process of demagnetising by reversals without the end 
connections being interfered with. The solenoid, which was wound on a 
brass tube, served the additional purpose of protecting the suspended 
system from air currents. The temperature of the room was kept as nearly 
as possible constant at 19*5° C. by means of several small gas jets. Ther- 
mometers placed at the upper and lower ends of the wire did not vary 
beyond the limits 19*3° C. and 19 7° C. The wire was placed in position 
five hours previous to readings being taken, a precaution which ensured 
that it had assumed a uniform temperature throughout. Determinations 
of the period of vibration before and after the demagnetising process 
showed no indication of change. If any alteration does take place, it is 
certainly less than 1 part in 4000. The slight variations in temperature 
did not affect the readings, as the consequent change in the rigidity 
modulus would not exceed 0’01 per cent., according to the observations of 
Katzenelsohn and Pisati. 
As it seemed possible that the molecular strains taking place at each 
swing of the vibrating system might have to a considerable degree 
removed the “ sensitive state ” early in the experiment, a static method of 
measuring the modulus was afterwards employed. A freshly annealed 
wire was placed as in the previous experiment, the vibrator having a 
surrounding zinc case of large radius. To the circumference of this case 
a finely graduated scale was attached, which could be observed by a 
reading telescope provided with cross wires, and placed at a distance of 
several feet from the wire. To the vibrator was attached a cylindrical 
brass rod of about 8 mms. diameter, the axis of which coincided with that 
of the test wire. Fine threads of equal weight were attached to this rod, 
given a few turns round it, passed over pulleys mounted on ball bearings, 
and attached at their free ends to 50-gramme weights. By this means a 
torsional couple was applied to the wire, and the amount of the resulting 
twist was measured by means of the telescope and scale. After demagnet- 
isation by reversals, no variation could be detected in the angle of twist 
produced by the same couple. Several other varieties of steel wire were 
tried with like result, although in some cases the sensitive state wiped 
out amounted to as much as 35 per cent, for a field strength of 8 to 12 c.G.s. 
units. 
Younq's Modulus . — Measurements of Young s modulus by the bending 
VOL. xxviii. 40 
