[ 441 ] 



XIX. Results of the Magnetic Ohservations at the Kew Observatory. — No. Ill, 

 By Lieut-General Edwaed Sabine, Ii.A., President oftlie Royal Society. 



Eeceived June 21,— Eead June 21, 1866. 



§ 9. Lunar-diurnal Variation of the three Magnetic Elements. 



The recognition of a cosmical origin of some of the variations of terrestrial magnetism 

 has made it desirable to employ in magnetic observatories apparatus of a more exact 

 and dependable character, and methods of dealing with the results thus obtained of a 

 more close and rigorous description, than were previously thought requisite. The pre- 

 sent communication is directed to the discussion of the Lunar-diurnal Variation of the 

 three magnetic elements shown by the instruments and methods adopted at the Kew 

 Obsei-vatory, commencing in 1858, and continued as far as the reductions have at pre- 

 sent proceeded, viz. to the close of 1864. It has the double purpose, first, of making 

 known the systematic and highly satisfactory character of the results which have been 

 already obtained ; and, second, of acting in some measure as a guide, and certainly as an 

 encouragement, to the several establishments at home and abroad which have adopted 

 the Kew System of magnetic investigation. 



The instruments employed for the determination of the lunar-diurnal variation furnish 

 a continuous photographic registry of the changes in the direction of a magnet whose 

 motion is limited to a horizontal plane, and in the amounts of the horizontal and vertical 

 components of the force acting on a freely suspended magnet. The photograms which 

 record these changes are submitted to a very careful and exact process, by which the 

 variations from a permanent zero, of the horizontal magnetic direction, and of the 

 amounts of the horizontal and vertical components of the Force, are measured and tabu- 

 lated at twenty-four equidistant intervals of an astronomical solar day. The accuracy 

 of the tabulations is checked by a repetition of the process of measurement by different 

 persons, and by a reexamination in cases of discrepancy. The proportion of failures in 

 the hourly records from any or from all causes whatsoever is very small — less in fact 

 than 1 per cent, throughout the whole period. The subjoined Table (No. I.) shows 

 that of 175344 positions which should have been recorded, there were only 1497 

 failures, of which 103 were occasioned by the employment of the instruments in other 

 experimental processes. It must be remembered also that the period under notice is 

 that of the commencement of the record, when experience has to be gained, and the 

 causes of accidental failures have to be remedied. 



The positions thus measured from the photograms at every hour of astronomical 

 time, and entered in monthly tables, were then subjected to the usual processes, first, 



MDCCCLXVI. 3 <t 



