MAGNETIC DECLINATION NEAR THE MAGNETIC EQUATOR. 679 



with the position of the moon. When we examine the differences projected under 

 the lunar hours, the law of lunar action is seen more clearly; and it appears with 

 much regularity, considering that irregular solar and other actions are included. 

 In order to diminish the effect of these irregularities, I have taken the means, and 

 projected — 1st, The mean of the first six curves, those for the days, December 16 d 

 8\1858, till 23 d 14\ when the moon had the greatest north declination; 2d, The 

 mean of the next five, when the moon was near the equator, going south (Es) ; 

 3d, Of the next six curves (Dec. 29 d 19 h till January 6 d l h ), when the moon was 

 farthest south ; 4th, Of the following five curves, when the moon was near the 

 equator, moving northwards (En) ; and, 5th, The mean of all the curves in the 

 lunation (see fifth division of Plate XLIII.) 



The latter presents two nearly equal maxima and two nearly equal minima ; 

 the former near the moon's superior and inferior passages of the meridian, and 

 the latter about six hours before and after. When the moon was farthest north, 

 the maximum near the inferior passage was the greater, the minimum near the 

 moon setting being most marked. With the moon near the equator going south, 

 the maxima were nearly equal, but the moon set minimum was by far the most 

 marked. When the moon was farthest south, the maximum near the superior 

 passage was the greater, and the minima were nearly equal. With the moon near 

 the equator going north, the maximum near the superior passage was the greater, 

 and the minimum near moon-rise by far the most marked. 



These results, derived from a single lunation, agree almost exactly with those 

 which I had previously deduced from six years' observations.* The dotted curve 

 for the moon near the equator, moving north, including the observations two days 

 later, or to 14th January, represents better the conclusions previously obtained for 

 this position of the moon, in having nearly equal maxima. 



These changes of the law of lunar diurnal variation of magnetic declina- 

 tion understood, it will be perceived that the daily curves (second and fourth 

 divisions of Plate XLIII.) in general follow the law and its changes with consider- 

 able fidelity. 



The range of the solar diurnal curve for December 1858 and January 1859 is 

 nearly 2' 2. The ranges of the mean curves (December 16th 1858, till January 

 12th 1859) are for the moon : — 



Farthest north, .....= 1'70 



Near the equator, going south, = 1/-95 



Farthest south, ..... = l' - 90 



Near the equator, going north, = 2'-70 



I believe that we may conclude, as I have ahead} 7 stated, that the effect of 

 the lunar action is sometimes greater than that of the solar action in the diurnal 



* See " Proceedings of the Royal Society of London,'' vol. x. pp. 482-3. 



