aming theories, does the imagination _create_ its
materials. It expands, diminishes, moulds, and refines, as the case
may be, materials derived from the world of fact and observation.

This is more evidently the case in a theory like that of light, where
the motions of a subsensible medium, the ether, are presented to the
mind. But no theory escapes the condition. Newton took care not to
encumber the idea of gravitation with unnecessary physical
conceptions; but we know that he indulged in them, though he did not
connect them with his theory. But even the theory, as it stands, did
not enter the mind as a revelation dissevered from the world of
experience. The germ of the conception that the sun and planets are
held together by a force of attraction is to be found in the fact that
a magnet had been previously seen to attract iron. The notion of
matter attracting matter came thus from without, not from within. In
our present lecture the magnetic force must serve as the portal into a
new domain; but in the first place we must master its elementary
phenomena.

The general facts of magnetism are most simply illustrated by a
magnetized bar of steel, commonly called a bar magnet. Placing such a
magnet upright upon a table, and bringing a magnetic needle near its
bottom, one end of the needle is observed to retreat from the magnet,
while the other as promptly approaches. The needle is held quivering
there by some invisible influence exerted upon it. Raising the needle
along the magnet, but still avoiding contact, the rapidity of its
oscillations decreases, because the force acting upon it becomes
weaker. At the centre the oscillations cease. Above the centre, the
end of the needle which had been previously drawn towards the magnet
retreats, and the opposite end approaches. As we ascend higher, the
oscillations become more violent, because the force becomes stronger.
At the upper end of the magnet, as at the lower, the force reaches a
maximum; but all the lower half of the magnet, from E to S (fig. 22),
attracts one end of the needle, while all the upper half, from E to N,
attracts the opposite end. This _doubleness_ of the magnetic force is
called _polarity_, and the points near the ends of the magnet in which
the forces seem concentrated are called its _poles_.

[Illustration: Fig. 22.]

What, then, will occur if we break this magnet in two at the centre E?
Shall we obtain two magnets, each with a single pole? No; each hal