OF THE DIP. 



19 



The diagram shows the hours of no semi-annual change as follows: 5| A.M.; 

 1 P. M.; 3 P. M., and 7 P. M.; greatest change at 10 A. M., secondary at 6 P. M., 

 with a range of 0.'51 and 0.'18, respectively. 



The turning epochs are found by the variation at the hour 10 A. M., when the 

 monthly differences from the annual mean are as foUows : — 



January .... 



— 0.'19 



July ..... 



+0.'28 



February .... 



—0.30 



August .... 



+0.50 



Marcli 



—0.24 



September . . . 



+0.33 



April 



+0.19 



October .... 



—0.08 



May 



+0.08 



NoTember . . . 



—0.25 



June 



+0.15 



December . . . 



—0.42 



These values are represented by the formula 



A,= 0.'35 sin {6 + 253° 38') + O.'IO sin (29 + 328° 39') 

 the angle 6 counting from January 1st. This formula gives a change of sign for 

 the middle of April and the middle of October, or about 25 days after the equinoxes. 



ANALYSIS OF THE SOLAR-DIURNAL VARIATION OF THE DIP. 



In the following formulae expressing the solar-diurnal variation of the dip for 

 each month, summer, winter, and year, the angle 6 counts from midnight at the 

 rate of 15° an hour; a positive sign indicates increase of north dip, a negative sign 

 the reverse. The expressions are derived directly from Table IV. 



For January 

 For February 



For Marcli A 



For April A 



For May A 



For June A 



For July A 



For August A 



For September a 



For October a 



For November A 



For December a 



= 0.'24 

 = 0.'34s. 



n (S + 284O19') + 0.'33s; 

 n (9 + 246O28') + 0.'30 



= 0.'40 sin (9 + 2510 24') + 0.'24 s 

 = 0.'55 sin (6 + 263° 10') + 0.'23 s; 

 = 0.'50 sin (9 + 2710 33') + 0.'22 s: 

 = 0.'53 sin (9 + 2740 37') + 0.'2] 

 = 0.'64 sin (9 + 2750 34') + 0.'24 s; 

 = 0.'68 sin (9 + 2720 55') + 0.'32 si 

 = 0.'70 sin (9 + 260O 14') + 0.'27 s; 

 = 0.'41 sin (9 + 2560 31') + 0.'27 s 

 = 0.'28 sin (9 + 2590 21') + 0.'27 

 = 0.>2a sin (9 + 2580 09') + 0.'37 



n (29+ 780 15') + 0.'14 Si 

 :» (29+ 750 40') + 0.'08 s; 

 n (29+ 850 27') + 0.'12 s 

 n (29 + 1090 36') + 0.'18s 

 n (29+146O40') + 0.'14s; 

 n (29 + 1090 03') + 0.'19 si 

 n (29 + 1310 08') + 0.'14s 

 ■n (29 + 132O56') + 0.'24s 

 :n (29 + 116O49') + 0.'23si 

 n (29 + 1010 19') +0.'ll Si 

 n (29+ 990 30') + 0.'08 s: 

 n (29+ 770 45') + 0.'12 s: 



n (39 + 2820) 

 n (39+3080) 

 n (39 + 303O) 

 n (39 + 330O) 

 n (39+ OO) 

 + 3310) 

 :« (39 + 3220) 

 n (39+3310) 

 n (39+3490) 

 n (39 + 3350) 

 n (39 + 3280) 

 n (39+2580) 



We have also:- 



For Summer 

 For Winter 

 For Year 



A; = 0.'60sm (9 + 269O37') + 0.'25 i 

 Aj = 0.'30sm (9 + 260O57') + 0.'29 . 



(29 + 1250 02') + O.'IS sin (39 + 3350) 

 (29 + 850 12') + O.'IO sin (39 + 300O) 



J + 2660 47') + o.'25 sin (29 + 103O 19') + O.'IS sin (39 + 3220) 



Summer is comprised of the months from April to September; winter of the 

 months from October to March. 



The following comparison of the observed and computed values for August shows 

 about average differences ; in general the summer values are a little better repre- 

 sented than the winter values. 



