(i I [' II E M A G N ET I C I'O i: (' E. 



23 



The preceding mean diurnal variations were put in the following analytical form, 



in which the angle counts from midnight at the rate of 15° an hour. 



Ve^r 1S4C-U II = 223>.5 + E .98 tin (9 + 252° 14') + 11«.68 tin (2 + 121° 16') + 5«.89 tin (3 9 + 314° 42') 



" 1841-12 H= 551.5 + 4.03 sin (fl -4- 244 07)+ 6.58 sin (2 8+131 32 ) + 4.48 sin (3 9+312 19) 



" 1842-43 H= 789.9 + 4.14 sin (0+260 06)+ "•"" "" (2 8 + 132 24 ) + 3.74 sin (3 8+323 06) 



" 1843-44 H = 1007.4 + 2.14 sin (9+273 55)+ 5.09 tin (2 9 + 128 58 ) + 2.35 sin (3 + 317 58) 



u is44_45 H = 1191.4 + 4.40 sin (0 + 271 13)+ 6.8G sin (2 + 123 25 ) + 4.11 sin (3 0+321 



To show the degree of correspondence in the formula 1 when deduced from the 



observations of the even and odd hours separately, the results for the last year have 



been added, viz: — 



Even hours II = 119V.3 + 4 1 .20 sin (0 + 271° 28') + 6 1 .98 sin (2 + 122' 360 + 4\11 sin (3 9 + 322' 35') 

 Odd hours H = 1191.5 + 4.60 tin (9 + 270 09 ) + 6.73 sin (2 0+124 13 ) + 4.12 sin (3 + 320 17 ) 



The close agreement between the observed and computed values is shown gene- 

 rally in the annexed diagram. 



(A). — Inequality id the Diurnal Variation of the Horizontal Intensity. 



204 div 

 08 



1840-41. 



1*41-42. 



1*42-43. 



1S43-44. 



1-44-45. 



0" 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 IS 19 20 21 22 23 24" 

 A. M. P- M- 



Philadelphia local time. 



The following table exhibits the differences for the year 184-2-43. as an example 

 of the numerical correspondence. 



