re 
PRESIDENTS ADDRESS—SECTION A. B15) 
have done, it would be still more improper to ignore the action 
of the material of magnets. Hopkinson has found that the 
magnetisation of iron is so nearly destroyed by a trace of 
manganese, that no hypothesis that we can frame of the distri- 
bution of the manganese as such through the iron, will 
account for the effect observed. Chemists will admit that in 
manganese iron there is probably some combination between the 
iron and manganese, and I think Hopkinson’s experiments prove 
not only that such must be the case, but that an amount of 
chemical combination, which may be described as very moderate, 
will entirely alter the magnetic behaviour of the iron. Conse- 
quently I think we may say that whatever be the cause of 
magnetisation, assuming it to be an ether phenomenon—that 
phenomenon is immensely influenced by small chemical change— 
a change for which the ether usually gets little credit. Hopkin- 
son’s experiments seem to force on us the conviction that if 
magnetic phenomena depend on the ether, then at least in iron 
the relation of the ether to the matter of the iron must be 
extremely close. We are almost driven to think that magnetism 
may be produced by some peculiar mode of motion of iron 
molecules which is shared to only a slight extent by the 
molecules of other materials. If we take this view, it seems 
to follow that the particular mode of molecular motion which 
‘enables the ether to give rise to magnetic effects must be very 
nicely balanced since it takes so little to upset it. 
We are now in a position to return to the final principle of 
Maxwell’s theory, which may be stated in a theorem analogous 
to that relating to the electromotive force produced by change 
of magnetic induction. It is that “the line integral of the 
magnetic intensity round any closed curve is equal to 4 7 times 
the current through the curve.” It need hardly be pointed out 
that this is experimentally true when the current is in a wire, 
and indeed it is the principle on which dynamos are designed. 
Maxwell’s extension amounts to assuming that it is also true, 
when the current is due to change of electric induction. It is 
obvious that there is a close reciprocal relation between this 
principle and that last discussed, and this is exceedingly well 
brought out by Poynting in his so-called modification of the 
principle. As has been said, it follows from the theory in 
general that the energy keeping a current going moves in 
sideways. Part of this energy may be considered to be carried 
by the tubes of electric induction, and an equal part by the 
motion of the tubes of magnetic induction, which by the principle 
just stated must accompany them. If we look on the tubes of 
electric induction as being in motion, then the principle may be 
stated as follows: Giving the name “magnetomotive force” to 
the line integral of magnetic intensity in accordance with a 
suggestion made by Bosanquet and adopted by Poynting, then 
c2 
