238 M. Werner Siemens on the 
have been based upon the first-known electrical phenomena. 
But permanent magnetism is only a secondary phenomenon. 
It is what remains of a previous more intense magnetization 
whose laws can only be deduced from those of electromag- 
netism, since magnetism is undoubtedly to be conceived only 
as an electrical phenomenon. ‘The electric current, or, more 
generally, electricity in motion, is the only known source of 
all magnetism. I have already expressed the opinion that 
this must also hold good of the earth’s magnetism, and have 
assigned as a reason that, so far at least, no other cause is 
imaginable than the rotation of the electricity accumulated at 
rest upon the earth’s surface with the earth about its axis. 
The loadstone, and other bodies occurring in nature in a 
magnetic condition, obviously derive their magnetism from 
that of the earth, or in particular cases possibly from the direct 
action of electric discharges. 
If, on the other hand, we start with the assumption of a 
body directly or indirectly magnetizable by means of electric 
currents, and which retains no magnetism so soon as the cause 
of magnetism ceases to act, and assume with Faraday that, 
moreover, the magnetic action, both in the magnetic body 
itself as well as in the surrounding space, is propagated only 
from molecule to molecule, or from space-element to space- 
element, then the further assumption lies close to our hand 
that both actions—the internal and external—must be com- 
pletely dependent on each other. In an iron rod round which 
an electric current circulates there can be, then, only so much 
magnetism produced by the electric disturbance which acts 
upon it as is associated in the space surrounding the iron rod 
with the magnetic disturbance—advancing in the direction of 
Haraday’s lines of force from the north magnetic to the south 
magnetic superficial elements of the rod—and thus arranged 
in the form of a magnetic closed circuit. 
If this conception is shown by experiment to be allow- 
able, then the laws of the molecular communication of heat, 
electricity, and electrostatic distribution must also hold good, 
with the necessary modification for magnetism. We should 
then be able to lay down a universal law for the intensity of 
magnetism of the form “Sum of the magnetizing forces di- 
vided by the sum of the opposed resistances,” which would 
avoid many difficulties and apparent contradictions. The 
following further law must also hold good, that “ In each 
sectional plane which cuts all the lines of force the sum of the 
magnetic moments of all the magnetic molecules cut through 
=0.’ Such a sectional plane can pass only through the 
magnetic centre of the magnetized body cutting through 
