mpted to collect and generalize observations on
the variation of the needle, was Robert Halley, who constructed a chart,
showing a series of lines drawn through the points or places where the
needle exhibited the same variation. This chart was published in 1700, and
was preceded by some exceedingly curious papers, communicated to the Royal
Society, in which he expresses his belief that he has put it past doubt
that the globe of the earth is one great magnet, having four magnetic
poles or points of attraction, two near each pole of the equator; and that
in those parts of the world which lie adjacent to any one of those
magnetical poles, the needle is chiefly governed thereby, the nearest pole
being always predominant over the more remote.

The great importance of collecting as much information as possible
respecting the laws of magnetism, with a view to the proper understanding
of its effects, was fully understood by Halley, as the following passage,
taken from one of his papers, read before the Royal Society in 1692,
singularly attests: "The nice determination of the variation, and several
other particulars in the magnetic system, is reserved for a remote
posterity. All that we can hope to do is, to leave behind us observations
that may be confided in, and to propose hypotheses which after-ages may
examine, amend, or refute; only here I must take leave to recommend to all
masters of ships, and all others, lovers of natural truths, that they use
their utmost diligence to make, or procure to be made, observations of
these variations in all parts of the world, as well in the north as south
latitude, after the laudable custom of our East India commanders; and that
they please to communicate them to the Royal Society, in order to leave as
complete a history as may be to those that are hereafter to compare all
together, and to complete and perfect this abstruse theory."

Halley's theory, or rather hypothesis, which regarded our globe as a great
piece of clockwork, by which the poles of an internal magnet were carried
round in a cycle of determinate but unknown period, was so far confirmed,
that his variation chart had been hardly forty years completed, when, by
the effect of these changes, it had already become obsolete; and to
satisfy the requirements of navigation, it became necessary to reconstruct
it. This was performed by the aid of various observations furnished by the
Commissioners of the Navy, and the East India, Afric