302 DK. CHARLES CHREE: SOME PHENOMENA OF 



The great reduction in the difference between day and night as we pass from quiet 

 to disturbed days seems noteworthy. The accordance between day and night in the 

 d q columns is remarkable. 



28. When dealing with the phenomena at a single station, the vertical is probably 

 the most convenient initial line. If, however, we wished to compare results at stations 

 in different latitudes, it would probably be found more convenient to define the 

 resultant force vector by the intensity R which is, of course, the same for all 

 co-ordinate systems the angle $ which R makes with the earth's axis, and the angle ^ 

 which a plane parallel to the earth's axis and containing R makes with the meridian 

 plane. 



If X denote the latitude of the station, and 6 and <f> have their previous meanings, 

 and if P and Q denote the components of II in the meridian plane, respectively parallel 

 and perpendicular to the earth's axis, then 



P = R (sin X cos cos X sin 6 cos </>), 

 Q = R (cos X cos #+sin X sin 9 cos <). 



The positive direction of P is here taken from the North to the South Pole, and 

 the positive direction of Q towards the earth's axis. The third rectangular co-ordinate 

 of R is AE (or -AW), i.e., R sin 6 sin <. 



The relations connecting i/; and -% with 6 and <f> are 



ijj = cos" 1 (sin X cos 8 cos X sin 9 cos <), 

 X = cot"" 1 (cos X cot 6 cosec ( + sin X cot </>). 

 At Kew, where X = 51 28' 6", these become 



ty = cos" 1 (782 cos #--623 sin 9 cos <), 



x = cot' 1 ("623 cot 9 cosec < + 782 cot <). 



Comparison of Diurnal Inequalities at Different Stations. 



29. The results arrived at in this and my previous papers suggest a number of 

 considerations relating to the tabulation and interpretation of magnetic curves. Some 

 stations in deriving diurnal inequalities limit themselves to a selection of the quieter 

 days. There may be only five of these a mouth, as at Kew and Falmouth, or 10, as 

 at the stations of the U.S. Coast and Geodetic Survey. Other stations derive diurnal 

 inequalities from all days, excepting as a rule those specially disturbed. A few 

 stations, e.g., Greenwich and Pawlowsk, publish regularly inequalities of two kinds, 

 viz., from quiet days alone the Pawlowsk quiet days are usually only two or three 

 a month and from all but highly disturbed days. At some stations the curves, even 

 those from quiet days, are systematically smoothed ; at others mean hourly values are 



