716 
PROFESSOR F. G. BAILY ON THE HYSTERESIS OF 
when but few combinations are dissolved and the movement is of a quasi-elastic 
nature, the amount of energy dissipated will not be greatly different from that 
dissipated in an alternating field of equal strength. In fact, since the molecular 
changes will be forced to take place in one direction only, there will be on the 
average a greater resistance to movement, and in consequence an increased dis¬ 
sipation of energy. When, however, the field becomes stronger, each molecule will 
develop a tendency to rotate in synchronism with the field and be less affected by 
the magnetic influences of surrounding molecules. At this point the hysteresis will 
show a considerable diminution which will become more marked as the field increases 
in strength, until finally every molecule will rotate in unison with the field with 
complete absence of oscillatory movement. The value of the hysteresis under these 
last conditions will furnish an important clue to the nature of hysteresis. If the 
hysteresis sensibly vanishes at this point it will be strong evidence that the damping 
of the movements of the molecules is not due even partially to mechanical friction, 
but must be produced by some action which is called into play by the rapid oscil¬ 
lations of the molecular magnets, but not by the comparatively slow motion of their 
rotation with the field. Since the difference between these speeds must be enormous, 
the damping may be due either to some form of eddy currents or possibly to some 
form of fluid friction. 
The first experimental work upon hysteresis in a rotating field was carried out by 
Professor Ferraris (‘ Atti d. P. Acc. di Torino,’ No. 23, 1888). Producing a rotating 
field by means of two coils at right angles supplied with alternating currents of 
approximately one quarter difference of phase, he showed that a laminated iron core 
would rotate by reason of its hysteresis. Beyond proving that at low speeds the 
hysteresis was independent of the speed, his results were not quantitative, owing to 
irregularities caused by vibration in the apparatus. 
In order to investigate the matter the following apparatus was designed. A 
powerful electro-magnet is caused to rotate on fixed spindles arranged so that the 
axis of rotation passes between the pole pieces, which are bored out to a cylindrical 
form concentric with the axis. A cylindrical armature is held on pivot bearings in 
the fixed spindles concentrically between the poles. The direction of magnetisation 
rotates in a plane at right angles to the axis of the armature and concentric with its 
axis. The armature, though free to rotate in its bearings, is prevented from 
continuous rotation by a spring attached at one end to its spindle, and at the other 
end to the fixed spindles of the magnet. Movement of the armature is indicated by 
a beam of light reflected from a small mirror attached to the armature on to a 
circular scale concentric with the armature axis. 
The apparatus is shown in fig. 1. The electro-magnet is of Swedish iron, 
8 sq. centims. in cross-section inside the coils. The pole pieces are bored out to a 
diameter of 2'3 centims., and subtend an angle of 120°. They are axially # of the 
same length as the armature. The excitation is produced by two coils each of 316 
