( 511.) 
XXUI.—On the Horizontal Force of the Earth’s Magnetism. By Jonn ALLAN 
Broun, F.R.S., Director of the Observatories of His Highness the Rajah of 
Travancore. (With Six Plates.) 
(Read 4th February 1861.) 
The only observations made with Dr Lioyn’s bifilar magnetometer, published 
with corrections applied to the individual observations, are those made in the 
Makerstoun Observatory, forming part of the Transactions of the Royal Society 
of Edinburgh. The results obtained from these observations (especially from 
those for 1844 and 1845), were first compared by me in 1856, with observations 
(also corrected by myself) made in the Trevandrum Observatory during the same 
years. The singular resemblance of the variations of daily mean intensity thus 
discovered at two places so distant, and so differently situated on the earth’s sur- 
face, induced me to undertake the considerable labour of determining the tem- 
perature coefficients, and of correcting and discussing all the published (and some 
unpublished) observations made in the colonial observatories. This labour was 
too great to have been undertaken by me alone, in consistence with my other 
duties, and it is due to the liberality of His Highness the Rajah of Travancore, 
that I could employ in part for this work the computers attached to his Observa- 
tory. The coefficients obtained for the colonial bifilars, by a method already 
described by me,* are given in a paper on the errors of the instrument.} The 
results deduced after corrections applied, will be found in the following paper; 
the troublesome relations and connections of broken series and corrections for 
_ accidental errors, discovered after a most careful examination of all the observa- 
tions, are given in an appendix, not to complicate the discussions. 
Yearly Mean Value of the Horizontal Intensity and iis Secular Variation. 
2. It has been supposed that the secular variation of the horizontal intensity 
could only be determined by absolute observations. The readings of the bifilar 
magnetometer being liable to so many accidental errors, they have been con- 
sidered valueless for this end. This conclusion, there can be no doubt, has had a 
considerable basis; yet it appears to me, that in many cases, the bifilar results 
may be employed with reference even to this question. The absolute intensity 
cannot be observed to the same degree of exactness as its variations by the bifilar, 
and this is a matter of much importance in determining the /aw of secular varia- 
tion. It has appeared to me desirable to compare the results for different places, 
and especially those for Makerstoun and Hobarton, which have seemed trust- 
worthy for this purpose. 
* Trans, Royal Soc. Ed., vol. xvi. p. 74, See also “ Introduction to Makerstoun Observations.” 
t See Trans. Royal Soc. Ed., vol. xxii. p. 467. 
VOL. XXH. PART IT. 6Q 
