TRANSACTIONS OF THE SECTIONS. yf 
On the Effect of Height on the Diurnal Variation of the Horizontal Comple- 
ment of the Magnetic Force. By J. A. Broun, F.R.S.E. 
On the Variation with Season of the Differences of the Mean Pressure at 
Greenwich and Makerstoun. By J. A. Broun, F.R.S.E. 
On Electrical Figures of Dust on Plate Glass. ByJ.A. Broun, F.R.S.E. 
On the Effect of Height in the Atmosphere on the Diurnal Variation of Mag- 
netic Declination. By J. A. Broun, F.R.S.E. 
I brought before this Section at the Oxford Meeting; some notice of a series of 
magnetical observations, made under my direction, at the expense of General Sir 
Thomas Brisbane, for the purpose of determining the effect of height in the atmo- 
sphere upon the diurnal variations of the magnetic declination and force. This 
question is one of the greatest importance, as to the seat of the disturbing causes 
~ which produce magnetic variations ; it is, however, one of the most difficult kind for 
a practical answer. Those acquainted with the working of magnetical instruments 
know how rare it is to get two instruments of the same kind, and in the same room, 
to continue, even for a short period, to tell the same story. If it is so in the same 
building, it is evidently a difficulty of no common kind to obtain consistent compa- 
rative results, when one of the instruments is placed on the summit of a mountain, 
frequently enyeloped in cloud, with no better cover than a rickety tent, which 
threatens to fly away with observers and instruments in eyery blast that sweeps the 
bare surface of the hill. 
The lower station occupied in this investigation was the Makerstoun Observatory, 
the upper station was the summit of the highest of the Cheviot hills, about eighteen 
miles E.S.E. of Makerstoun, and 2656 feet above the level of the sea, or 2440 
feet, nearly half a mile, above Makerstoun. The stations were in sight of each 
- other. On the first occasion, in the beginning of June 1847, the same tent was 
made to serve on Cheviot for observatory and residence. The frequent threats of the 
wind, the shaking of the ground by the motion of the tent poles and accidental 
touches of the instrument, from the proximity of the observers, rendered several of 
the few days’ observations obtained of indifferent value; and I accordingly again 
went to Cheviot after the meeting of the Association in the end of August 1847. On 
this occasion I separated the sleeping apartment from the observatory, and rendered 
the latter, to a considerable extent, independent of any gale that might blow. The top 
of the hill has a covering of peat moss, in some places upwards of six feet thick. I 
had a trench of sufficient width cut into a bank of this moss, where it was about 
four feet thick, clearing out the moss till the rock was rendered bare. The remain- 
ing height of wall was obtained by large peat slabs, and the whole was roofed by 
means of a wooden ridge pole and thick tent cloth. This observatory was very steady, 
rather damp of course, but little else could be expected where thick clouds rest for 
many days together, and almost always during the night. The instruments placed 
in this observatory, were two portable magnetometers, the loan of which I owed to 
the kindness of Mr. Jones of London, the maker, and a barometer; there were 
thermometers outside. It is the results from the declination magnetometer to which 
I wish to draw the attention of the Section at present. ; This instrument was placed 
on a firmly braced tripod ; the value of the scale divisions was 1.444, determined with 
the greatest accuracy, by means of the horizontal circle attached to the instrument ; 
the estimations were made by tenths of ἃ scale division or 014: as an error of 051 
in the mean of two readings was almost certain, this scale was evidently too small : 
in the mean of several observations, however, the probable error becomes very small. 
The observatory declinometer was employed at Makerstoun, the scale of which can 
be estimated with much certainty to half a tenth of a minute. The observations 
were made at both places at two minutes, and at one minute before each hour, of 
Gottingen mean time at the hour, and one minute after it; there were thus four 
comparative observations obtained ateach hour. On the 27th and 28th of August, the 
