'The Transactions of the Bombay Geographical Society. From January 1863 to December 1864. (Edited by the Secretary.) Volume XVII.' [‎198] (369/524)

198
EFFECT OF THE ATMOSPHERE, &C.
In 1842 I first proposed to determine the temperature co-efficient,
from the usual observations, by a method which was published first in
the Edinburgh Transactions for 1845, and at a Meeting of the British
Association at Cambridge in the same year. My method was met at
the meeting by an objection from the Astronomer Royal that it as
sumed that the magnet was affected only by its own temperature, and not
by the temperature of the air or of the soil. As my method gave a
verv different co-efficient from that derived by the usual method, the
conclusion might have been easily accepted that the air had an effect,
and that the difference of the co-efficients found by the two methods
was due to this cause. And as the old method was continued by every
one, excepting myself, till very recently, there was some ground for
believing that the conclusion was generally accepted.
I stated in reply to the Astronomer Royal, that I had made the
experiments already referred to with artificial heat, proving that the
external temperature had nothing to do with the co-efficient, and I
noticed as a probable cause of the difference, the effect of heat on the
torsion force of the suspending wires. My researches since then have
shown this cause to be probably the true one : the difference also by the
two methods is variable with the instrument, that is to say, with the
twist of the particular wires, and it has been found in one case to be
nothing.*
I may notice, in conclusion, another result of the Bombay Observations
which appears in the introduction to several volumes, namely, a sup
posed effect of solar white light upon the Balance needle. (See Bombay
Observations for 1845, Introduction p. iv.) Professor Orlebar found the
moonlight had no effect, but it was found that lamplight had a similar
effect to solar light though to a less degree. (Bombay Observations
1847, Introduction p. xxvi.) It is only necessary to say that the effects
observed were chiefly due to currents of air generated by the heat rays
from the sun and lamp, which could not have been perceived had the
needle been in a vacuum : the non-effect of the moonlight is thus quite
explicable.
* So Dr. Lloyd has informed me in the case of the Dublin bifilar.