MAGNETIC DECLINATION AT TREVANDRUM. 749 



sphere, within an hour and a half of the moon's passages of the meridian. We 

 see from the present discussion how these means include different causes 

 of variation, producing absolute inversion of the laws (at least during the day) 

 near the magnetic equator, and that in such positions the mean for the year 

 may not represent the fact at any period ; also for different latitudes, it is pro- 

 bable that the law will vary during the year with the different lengths of the 

 day and night. 



37. It has long been known that the solar diurnal variation was greatest 

 during the day ; we now find this is true, also, for the lunar diurnal variation. 

 We might suppose that there is a greater amount of the magnetic ether heaped 

 up over the face of the earth next the sun than on the other side in the earth's 

 shadow, and that the moon's action upon this ether is thus greatest where the 

 greatest disturbance can be produced. In whatever direction the hypothetical 

 electrical currents proceed, their intensity (or quantity) diminishes within the 

 earth's shadow, and is it not improbable that the result now found for the varia- 

 tion of the easterly magnetic force will be found true also for the northerly 

 and vertical forces.""" 



Diurnal Variation ivith reference to the Moon in the Half Orbits farthest from 



and nearest to the Earth. 



38. The only other discussion in connection with the lunar diurnal variation 

 which it seems to me of sufficient importance to give here, is that connected 

 with the moon's distance from the earth. For this question the discussion was 

 divided into two parts : First, for the months October to April ; and second, 

 for the group May to September ; in each case the law of diurnal variation 

 within the group is sufficiently constant to allow a conclusion from the mean 

 variations. 



The equations of sines computed from the means are as follows : — 



October to April. 



r y =■ + O'-OO14cos0 - O'-O249sin0 + O"lO49cos20 - 0"0061 sin 20 | 

 Apogee J = + Q ,. 0249 gin ^ + 176 o 52 ,^ + ,. 1050 S i n ( 2 $ + 93° 19') jProb. error = 0'-005. 



( y = - O'-OO24cos0 - O'-O316sin0 + O'-1114cos20 - O'-O494sin20 ^ 

 Perigee j = + /. 0317sin ^ + i82°2i') + 0'-1218 sin (20 + 113° 540 / » °' 013 - 



May to September. 



<y = - O'-OO98cos0 + O'-O118sin0 - O'-O446cos20 + O'-OO46sin20 j 

 Apogee | ^ +O'-O153sin(0 + 320°4') + O'-O448sin(20 + 275°57') $ " = '' 013 - 



ty = ■+ O'O118cos0 + O'-OOO6sin0- O'-O578cos20 + O'-O236sin20 | 

 Perigee J = + '-0118sin (0 + 86° 53') + O'-O624sin(20 + 292°13') j " ~ 



"' The calculations are being made with, reference to these components of force. 



