'The Transactions of the Bombay Geographical Society. From January 1863 to December 1864. (Edited by the Secretary.) Volume XVII.' [‎195] (366/524)

MAGNETIC INTENSITY OF THE EARTH. 195
Fahr. But in the subsequent volumes the temperature co-efficient is
stated to be only 2*0 scale divisions ; so that on this authority alone
the quantities for 1847, which were discussed, were over corrected by a
co-efficient of 12 scale divisions. As the result of an over correction
is to cause the bifilar readings to increase with increasing temperature,
we see at once why the three tables employed show an apparent rela
tion betwixt increase of temperature and an increase of the earth’s
magnetic force.
It was assumed that the correction for the temperature of the mag
net was accurate, whereas the introduction to the volume itself showed
that the correction could not be depended on (Introduction 1848,
p. xvi.), and subsequent volumes proved it to be altogether erroneous.
So far the tables are quite untrustworthy for any deductions.
In the next place, it was also assumed that the earth’s magnetic
force did not vary in any degree similarly with the variation of the
temperature of the eir; indeed, it was assumed that the former scarcely
varied at all, but that the annual and diurnal variations were chiefly
due to the varying temperature of the air three feet above the ground,
and upon these hypotheses the co-efficient for the “ circumambient air”
was deduced.
It must however be evident, that if the annual or diurnal laws of the
magnetic force were in any degree similar to those of atmospheric temper
ature, it would be hasty to assume that the one was in any degree a cause
of the other. For example, in India the horizontal magnetic force in
creases from the morning till 11 a.m. or noon, and then decreases till near
midnight; this is somewhat like the law of the temperature variation,
but it would be opposed to all sound induction to assume on this ground
alone, that the increase of bifilar scale reading till II a. m. was due to
the increase of temperature till the same hour. And this not merely
because the bifilar reading diminishes till 1 or 2 p. m., while the temper
ature is still rising, but also and particularly because the temperature
law is similar for all latitudes and seasons, whereas the bifilar law varies
with both season and latitude. In Europe the minimum horizontal
magnetic force occurs near 10 a. m., —nearly the hour of the maximum
in India.
If then the diurnal variation of the bifilar in India, even after an
accurate correction for the effect of temperature upon the magnet, were