lso, that somewhat similar
movements are always present in a magnetic field, and though we do not
know how to make them close up in the proper way, it does not follow
that it is impossible for them to do so.

The bearing of all this upon the problem of the transmutation of
elements is evident. No one now will venture to deny its possibility as
strongly as it was denied a generation ago. It will also lead one to be
less confident in the theory that matter is indestructible. Assuming the
vortex-ring theory of atoms to be true, if in any way such a ring could
be cut or broken, there would not remain two or more fragments of a ring
or atom. The whole would at once be dissolved into the ether. The ring
and rotary energy that made it an atom would be destroyed, but not the
substance it was made of, nor the energy which was embodied therein. For
a long time philosophers have argued, and commonsense has agreed with
them, that an atom which could not be ideally broken into two parts was
impossible, that one could at any rate think of half an atom as a real
objective possibility. This vortex-ring theory shows easily how possible
it is to-day to think what once was philosophically incredible. It shows
that metaphysical reasoning may be ever so clear and apparently
irrefragable, yet for all that it may be very unsound. The trouble does
not come so much from the logic as from the assumption upon which the
logic is founded. In this particular case the assumption was that the
ultimate particles of matter were hard, irrefragable somethings, without
necessary relations to anything else, or to energy, and irrefragable
only because no means had been found of breaking them.

The destructibility or indestructibility of the ether cannot be
considered from the same standpoint as that for matter, either ideally
or really. Not ideally, because we are utterly without any mechanical
conceptions of the substance upon which one can base either reason or
analogy; and not really, because we have no experimental evidence as to
its nature or mode of operation. If it be continuous, there are no
interspaces, and if it be illimitable there is no unfilled space
anywhere. Furthermore, one might infer that if in any way a portion of
the ether could be annihilated, what was left would at once fill up the
vacated space, so there would be no record left of what had happened.
Apparently, its destruction would be the destruction of a substance,
which is a very d