407 
The author concludes by a notice of the accidental errors to 
which the instrument is liable. 
3. On the Horizontal Force of the Earth’s Magnetism. By- 
John Allan Broun, F.R.S., Director of the Observatory of 
the Eajah of Travancore. Communicated by Professor Tait. 
The conclusions of this paper are derived from observations made 
at Makerstoun in Scotland ; Toronto, Canada ; Trevandrum, Travan- 
core, and Singapore, East Indies ; St Helena, the Cape of Good 
Hope, Hobarton, Van Dieman’s Island, &c. The author has cor- 
rected and discussed all the observations published made in the 
colonial observatories. 
The observations have been found affected by different errors, and 
frequently the series have been broken. An attempt has been 
made to correct these series ; and, in general, to render the series as 
near the truth as possible. The grounds for all these corrections are 
given in an appendix.* 
The results obtained may be summed up as follows : — 
Secular Variation and Mean Yearly Force. 
ls£, The secular change of the horizontal intensity followed the 
same law, and had the same value very nearly, at Makerstoun in 
Scotland, and at Hobarton, Van Dieman’s Island, the units being 
the absolute force at the respective places. 
2d, The law of change is, however, not linear, but had a maximum 
at both places, at the mean epoch, January 1844, and a minimum 
about July 1847. 
3d, At other stations the secular change has not always the 
same value as at Makerstoun and Hobarton, nor has it always the 
same duration. 
4f/i, The yearly mean force was sometimes an increase at all 
the stations, and it is concluded that the yearly mean intensity of 
the magnetism of the whole earth is not a constant quantity. 
5th, The secular change is probably a result of two causes — one 
dependent on the position of the earth’s magnetic poles, the other 
dependent on their varying intensity. 
* Some observations examined were found valueless, on account of errors 
which could not be corrected; and in one case the requisite data for accurate 
correction were wanting. 
