unless we are to regard the lines as simply indicating a sort of
atmosphere of magnetic effect whose density becomes less as we proceed
outward from the wire, in which case either form of symbol suffices.
The direction of polarization of the lines may be indicated by an
arrow head pointing in a direction of right-handed rotation in the
path of the lines. This is the typical figure or expression for all
forms of simple magnetic circuit--the form of the lines, their length,
position, density, will depend on the shape of the conductor or
conductors (when more than one) and the materials surrounding or in
proximity to the wire or wires.

If the current traversing the conductor is constant, the magnetic
field around it is stable and static, unless other influences come in
to modify it. The cutting off of the current is followed by
instability of the field whereby it can and must produce dynamic
effects. I say _must_ because the field represents stored energy, and
in disappearing _must_ give out that energy. To throw light on this
part of the subject is one of the objects of the present paper.
Cutting off the current supply in the case assumed leaves the
developed magnetic lines or strains unsupported. They at once shorten
their paths or circuits, collapsing upon the conductor as it were, and
continuing this action, cut the section of the conductor, and
apparently disappear in magnetic closed circuits of infinitesimal
diameter but of great strength of polarization. It appears to me that
we must either be prepared to give up the idea of lines of force or
take the position that the magnetic circuits precipitate themselves in
shortening their circuits and disappearing upon and cut the conductor.
It was Hughes who put forward the idea that an iron bar in losing its
apparent magnetism really short-circuits the lines in itself as
innumerable strongly magnetized closed circuits among the molecules.
In becoming magnetic once more these short circuits are opened or
extended into the air by some source of energy applied to strain the
lines, such as a current in a conductor around the bar.

May not this idea be extended, then, to include the magnetic medium,
the ether itself? Does it contain intensely polarized closed circuits
of magnetism which are ready to be stretched or extended under certain
conditions by the application of energy, which energy is returned by
the collapse of the extended circuits? This is doubtless but a crude