On the Diurnal Variations of Magnetic Elements. 37 



Smith showed that there existed a periodical difference in the 

 diurnal inequality, similar in character, and generally also in 

 magnitude, in each of the six years treated. Mr. Whipple, in 

 the British Association Report before alluded to, had previously 

 pointed out the existence of this periodical difference in the 

 three years with which he dealt, and suggested that it might 

 be in some measure due to instrumental causes, and in some 

 degree also to difference of geographical position, which con- 

 siderations would apply equally to the comparison of the 

 years 1883, 1886, and 1887; but how far such causes may 

 have acted, it was impossible from a comparison of this kind 

 to determine. 



Circumstances have now, however, arisen which enable us 

 to make a direct comparison of the two methods of tabulation. 



In compliance with a request made by the Sub-Committee 

 of the Kew Committee of the Royal Society, the Astronomer 

 Royal has arranged to select, yearly, five quiet days in each 

 month, to be generally adopted for the discussion and com- 

 parison of the diurnal inequalities of magnetism at the various 

 British magnetic observatories, in consequence of which 

 diurnal inequalities will now be determined at Greenwich for 

 these selected quiet days, as well as for all days in the manner 

 which has been so long employed. The w T ork at Greenwich 

 for the first year under this arrangement, that of 1889, 

 having been lately completed, opportunity is now given 

 for making a direct comparison of methods, a comparison in 

 which the question of difference of instruments or of geogra- 

 phical position in no way enters. Viewing the interest 

 attaching to this matter, the Astronomer Royal has kindly 

 given me permission to communicate to the Philosophical 

 Magazine the results arrived at for the year mentioned. 



The present comparison has been made not only for decli- 

 nation, but also for the horizontal and vertical forces, and the 

 results are contained in the accompanying Table. The diurnal 

 inequalities on the two systems, and their differences, are 

 given, in the case of declination in minutes of arc, and for the 

 horizontal and vertical forces in parts of these forces. But 

 in order to make the comparison between the several ele- 

 ments more complete, the differences have been further 

 converted into metric measure, those of declination being 

 expressed as westerly force, the unit employed in the Table 

 being '00001 of the metric unit. The metric differences are 

 also exhibited in the form of curves, for comparison with 

 the curves for declination given by Messrs. Robson and 



