ectro-magnetic
properties of metalliferous veins, some trace of electric currents seems
to have been detected in the interior of the earth.[756]

Some philosophers ascribe these currents to the chemical action going on
in the superficial parts of the globe to which air and water have the
readiest access; while others refer them, in part at least, to
thermo-electricity excited by the solar rays on the surface of the earth
during its rotation; successive parts of the atmosphere, land, and sea
being exposed to the influence of the sun, and then cooled again in the
night. That this idea is not a mere speculation, is proved by the
correspondence of the diurnal variations of the magnet with the apparent
motion of the sun; and by the greater amount of variation in summer than
in winter, and during the day than in the night. M. de la Rive, although
conceding that such minor variations of the needle may be due to
thermo-electricity, contends that the general phenomena of terrestrial
magnetism must be attributed to currents far more intense; which, though
liable to secular fluctuations, act with much greater constancy and
regularity than the causes which produce the diurnal variations.[757]
The remark seems just; yet it is difficult to assign limits to the
accumulated influence even of a very feeble force constantly acting on
the whole surface of the earth. This subject, however, must evidently
remain obscure, until we become acquainted with the causes which give a
determinate direction to the supposed electric currents. Already the
experiments of Faraday on the rotation of magnets have led him to
speculate on the manner in which the earth, when once it had become
magnetic, might produce electric currents within itself, in consequence
of its diurnal rotation.[758] We have seen also in a former chapter (p.
129) that the recent observations of Schwabe, 1852, have led Col. Sabine
to the discovery of a connection between certain periodical changes,
which take place in the spots on the sun, and a certain cycle of
variations in terrestrial magnetism. These seem to point to the
existence of a solar magnetic period, and suggest the idea of the sun's
magnetism exerting an influence on the mass of our planet.

In regard to thermo-electricity, I may remark, that it may be generated
by great inequalities of temperature, arising from a partial
distribution of volcanic heat. Wherever, for example, masses of rock
occur of great horizontal exte