ATOMICITY IN ELECTRICITY 
33 
^to that 
of cm 
t inc ^: 
Piassi 
^et as m,'j' 
T P iec e of the 1® * 
-«•ignetization is i" 
’ bQt °f molecnles- ■ 
2° «a***; 
bem ®tW0,Sfl^ 
P*" "There©; 
t 1 * *^§netiziiig iiua i 
stianoli to the i 
htse molecules®!' 
md that the; 
fe molecules so tiat 
Q me direction, ii. 
te modern theoiyi 
nted out byti 
molecnles do notti 
of the magnet^ 
cted on by a Iorcei 
, or because an ep 
)n of the entirety! 
mple expression i 
en of in the last ! 
arising from the: 
e with the delta 
re D denotes a® 
ecuiar directive t® 
npère, the map® 
irrent circulatiiigii: 
rular currents is 
rv to suppose that: 
i the current, 
1 of currents was ¡a 
ned the phenoms- 
d in the molecnle' 
? " molecular mf' 
Veber elements, ¡- 
term suggests dm 
lemical molecule. • 
w„ voL 87 1 p- ^ " 
III. 3] 
is obviously not the case. In the older theories of paramagnetism 
the Weber elements were supposed to turn under the influence of 
the applied field so as to set their axes in the direction of the field, 
and this tendency was supposed to be resisted by forces of a 
quasi-elastic nature. Weber also introduced forces of a frictional 
character to enable him to account for ferromagnetic effects. 
It is not a little remarkable that in papers written in 1871 
Wilhelm Weber pictured the hypothetical molecular current, 
which Ampère had imagined to be continually flowing inside 
molecules, as the rotation of light positive charges about heavy 
negative ones. He wrote as follows : “ The relation of the 
two particles as regards their motions is determined by the 
ratio of their masses e and e', on the assumption that in e and 
e' are included the masses of the ponderable atoms which are 
attached to the electrical atoms. Let e be the positive electrical 
particle ; let the negative be exactly equal and opposite, and 
therefore denoted by — e (instead of e'). But let a ponderable 
atom be attracted to the latter so that its mass is thereby so 
greatly increased as to make the mass of the positive particle 
vanishingly small in comparison. The particle — e may then 
be thought of as at rest, and the particle + e as in motion about 
the particle — e. The two unlike particles in the condition 
described constitute, then, an Ampèrian molecular current.” 
Weber’s model is identical with that assumed in the modern 
electron theory save in the fact that it is now the negative 
particle whose mass is negligible in comparison with that of 
the positive, so that the roles of the particles are reversed. 
Langevin has developed this electronic theory of magnetism in 
which a negative electron circles in an orbit about a positively 
charged nucleus. 
But to return to Weber’s work on paramagnetism and ferro 
magnetism. Sir Alfred Ewing showed that Weber’s assumption 
of frictional forces was unnecessary. He constructed models 
in which the controlling forces are magnetic only, and arise from 
the action of the neighbouring molecular magnets. This theory, 
put forward by Ewing in 1890, gave a qualitative explanation 
of the principal phenomena of ferromagnetism. It explained 
the three characteristic stages of magnetization as saturation is 
approached (Fig. 4). 
In the first stage the intensity of magnetization, I, increases 
slowly and in proportion to the magnetizing field, H ; in the 
second stage the curve showing the relation between I and H is 
very steep, the value of I rising rapidly as H increases. After 
this the increase in I is slow, and eventually as magnetic satura 
tion is approached I becomes nearly constant. 
When the Weber elements are caused to turn by an applied