e, based on the
mechanical conception of a limit to the possible vibrations of a vortex
ring, not to mention other more or less fascinating speculations based
on the vortex hypothesis, must be regarded, whatever their intrinsic
interest, as insecurely grounded, until such time as new experimental
methods shall give them another footing. Lord Kelvin himself holds all
such speculations utterly in abeyance. "The vortex theory," he says,
"is only a dream. Itself unproven, it can prove nothing, and any
speculations founded upon it are mere dreams about a dream."*1*

That certainly must be considered an unduly modest pronouncement
regarding the only workable hypothesis of the constitution of matter
that has ever been imagined; yet the fact certainly holds that the
vortex theory, the great contribution of the nineteenth century towards
the solution of a world-old problem, has not been carried beyond
the stage of hypothesis, and must be passed on, with its burden of
interesting corollaries, to another generation for the experimental
evidence that will lead to its acceptance or its refutation. Our century
has given experimental proof of the existence of the atom, but has not
been able to fathom in the same way the exact form or nature of this
ultimate particle of matter.

Equally in the dark are we as to the explanation of that strange
affinity for its neighbors which every atom manifests in some degree.
If we assume that the power which holds one atom to another is the same
which in the case of larger bodies we term gravitation, that answer
carries us but a little way, since, as we have seen, gravitation itself
is the greatest of mysteries. But again, how chances it that different
atoms attract one another in such varying degrees, so that, for example,
fluorine unites with everything it touches, argon with nothing? And how
is it that different kinds of atoms can hold to themselves such varying
numbers of fellow-atoms--oxygen one, hydrogen two, and so on? These
are questions for the future. The wisest chemist does not know why the
simplest chemical experiment results as it does. Take, for example, a
water-like solution of nitrate of silver, and let fall into it a few
drops of another water-like solution of hydrochloric acid; a white
insoluble precipitate of chloride of silver is formed. Any tyro in
chemistry could have predicted the result with absolute certainty. But
the prediction would have been based purely upon previous em