MAGNETIC DECLINATION" AT TREVANDRUM. 755 



of the greatest lunar action), in the years from January 1856 to January 1865 

 were taken (with the exception of that having new moon the 16th December 

 1857, large disturbances having occurred on the 17th, 18th, and 19th). 



The moon's hour angle corresponding to 6 h. 28 m. a.m. (mean time), was first 

 computed to the nearest minute for each of the days from new moon to the first 

 quarter, for each of the seventeen periods under discussion. The days on 

 which the moon's hour angle was between 18 # 5 h. and 175 h. (at the above 

 hour) were marked as Id.; those in which the hour angles were betwixt 

 17*5 h. and 16'5 h. as 2d.; and so on to 125 h. and 11 5 h. The variations 

 due to the lunar action were now arranged for Id. under each other, according 

 to the solar hours (beginning with 12 p.m.), so that all the observations made at 

 the sunrise hour (18^) were under each other; the means of the vertical 

 columns were then taken, and that at 18^ h. corresponded to the mean of the 

 calculated hour angles of the moon (= 18"06 h.) A similar calculation was 

 made for each of the following days for which the moon's mean hour angle differed 

 by an hour at 18^ h. mean solar time. In this way any possible error is due to the 

 variation of the moon's hour angle (differing on the average a quarter of an 

 hour from the mean) corresponding to 18^ h., and this error the previous results 

 have shown to be of less importance than that clue to overlapping of the obser- 

 vations before and after sunrise. The results of these combinations are given in 

 Table II., and they are projected in the third line of curves of Plate XX VIII., * 

 the observation near sunrise (6^ a.m.) having an O round the point, and that 

 near sunset a X . 



55. In this more accurate combination, we find that the needle's movement 

 westwards on the first and second days ends at sunrise; on the third day, the 

 greater part of the movement westwards takes place after sunrise ; on the fourth 

 and following days, the movement westward begins at sunrise. 



56. There is here, however, an additional fact of importance which presents 

 itself on the fourth day : the movement is diminished so much that the variation 

 has the appearance of being nearly obliterated. The moon's hour angle on this 

 day at the sunrise observation is 15 h., and it seems as if, in this middle position, 

 the needle were solicited by opposing forces. 



57. The fact then is established, that the direction of the movement changes 

 at or near sunrise, and that near the time when the needle emerges from the 

 earth's shadow into sunlight the moon's action is reversed. This sudden 

 reversal bears some resemblance to that shown in the curves from April to May, 

 or May to June; for example, in the curves for the first quarter (Plate XXVII.) 



* It must be remembered that these curves cannot be compared with those preceding them, day 

 by day; Id. (under Od. in the plate) does not correspond to new moon (but to half a day or more after 

 it), and there are only six days here included between the day for the moon on the meridian near 

 noon, and the moon on the meridian at 6 h. p.m., whereas seven days were included by the other method. 



VOL. XXVI. PART IV. 9 L 



