568 On the Variation of the Earth's Magnetic Force. [Feb. 2%, 



believed to be free from all error. The unit- coefficients were determined 

 with the greatest accuracy for both the instrument on the peak and that 

 on the plain, A^erified by different methods, giving results which did not 

 differ by one five-hundredth of the whole value. The temperature- 

 coefficients were also found with much exactness ; and, as hard steel 

 magnets were chosen expressly for these instruments, the temperature- 

 coefficients were small. 



The following results from the hourly observations, made during the 

 last four months of 1856, will be sufficient for my present object. 

 Taking the observations at the hours nearest to those for which Mr. 

 C. Chambers has obtained his result (namely, 0'' 30°^ and 2^ 30"" p.m.), I 

 find the mean change of horizontal force from the former to the latter 

 tim-e~ 



On the plain, 200 feet above the sea = - 0-0009760 X, 



On the mountain, 6200 feet above the sea = ~ 0-0009724 X ; 



so that the change on the mountain-peak was less than on the 

 plain by one two-hundred-and-seventieth. In each of the months 

 October and November the change was exactly the same at the tv^^o 



If we take double the interval, so as to keep the same hours in the 

 middle, I find the mean change from ll'^ 30^-^ a.m. to 3^ 30'^^ p.m.— 



The change was therefore one three-hundred-and-sixtieth less on the 

 peak than on the plain. The mean of the diurnal ranges for the four 

 months was 0-00215 X. 



It be seen that in the interval of four hours the change was as 

 great as the whole mean diurnal range at Bombay ; and if the quantities 

 had been given to five places of decimals only as for Bombay, the move- 

 ments would have -appeared exactly the same at the two stations. 



It will thus be seen that instead of -J- less for a difference of 32 feet in 

 height, I do not find more than for a difference of 6000 feet when 

 the change during four hours is considered ; nor, till I have made a more 

 searching investigation of the whole series of observations, can I vouch 

 that this difference (which is very much less than the probable error of an 

 observation at either station) is not accidental. 



Another series of observations was made at the same two stations in 

 1861, when a unifilar horizontal-force magnetometer, on Dr. Lament's 

 construction, was employed, as well as the bifilar instrument. These 

 observations are not yet completely discussed ; and on that account I do 

 not enter at present into the question as to what difference may exist in 

 the laws of magnetic variations when the height differs by 6000 feet. 

 Meanwhile it will be useful, I believe, to those who may attempt investi- 



statious. 



On the plain, 200 feet above the sea 

 On the peak, 6200 feet above the sea 



= - 0-0016556 X, 

 = - 0-0016510 X, 



