distinct, and sometimes united with each
other in various but regular proportions; and these capacities of
coalescing with one class of bodies, and of remaining unaffected by
another, are called chemical "affinities." This is a convenient
generalization, and has properly received a specific name; though the
common appellation throws no light on the _cause_ of the phenomena,
which remains an impenetrable secret. To say that certain action is
_caused_ by the operation of chemical affinities is only to arrange it
with a large class of other observed appearances, equally obscure as to
their origin and essential character.

Let us go a step further, and suppose that the progress of discovery has
made known certain facts lying behind the phenomena in question, to
which they may all be referred. Let us suppose, that all bodies which
gravitate towards each other are found to be embosomed in a subtile,
ambient fluid, which connects them, as it were, into one system; that
the positive and negative states of electricity are resolvable into the
presence of two fluids standing in certain relations to each other; and
that substances show chemical affinity for each other only when they are
in opposite electrical conditions. Still, we have only advanced a step
in the generalization, and the real, efficient _cause_ of the
appearances is still hidden from us by an impenetrable veil. Gravitation
is now referred to the communication of motion by impulse; electricity,
to the combination and separation of different fluids; affinity, to the
attraction or repulsion of these fluids. The latter classes of phenomena
are more general, but not a whit more explicable, than the former. We
have now fewer causes to seek for, but not one of these few has been
discovered. When we have resolved electricity or gravitation into the
presence of an elastic medium, it is a mere figure of speech to say,
that we have discovered the _cause_ of the electric phenomena or of
gravity. That is just as far off as ever; for we have yet to discover
the principle whence flow _necessarily_ all the phenomena observable in
fluids. It is the sole end and the highest ambition of science to
discover as many as possible of the relationships which bind facts
together, and thus to carry the generalization to the farthest point.
Its office is not to discover causes, but to generalize effects. The
investigation of real causes is quite given up, as a hopeless
undertaking.

Observe