38 



RESULTS AND DISCUSSION OF THE 



The index error to needle No. 2 we find by direct comparison with needle No. I 

 on the following dates : 



.... correction -)-6.'0 



" +3.8 

 . . . " +1.0 



" 4.6 

 . . . . " +2.3 . 



17, 1842 

 11, 1842 



18, 1843 

 24, 1843 



Mean +1.'7 



May 



October 



July 



August 



September 12, 1843 



The correction to the Lloyd needles No. 1 and No. 3, A end north, we obtain also 

 by direct comparison, viz : 



August 29, 1843, correction to Lloyd No. 1, +4.'0 to Lloyd No. 3, +10.'0 

 September 5, 1843, " " " +0.7 " +10.3 



September 12, 1853, " " " +1.1 " " +9.8 



Mean correction, +1.9 " +10.00 



The corrections for polarity to needles 1 and 2 in 1843, are determined as follows: 



For needle No. 1. Mean dip in 1843 from 34 results, A north, 71 62/4 

 " " " " " B north, 71 53.8 



Mean dip 71 58.1 



Hence correction to needle 1, A north, 4/3, and B north + 4/3 



For needle No. 2 we have the following differences : 



May 17, 1842, A north B north . . . 8/4 1 



October 11, 1842, " " ... 11.9 



July 18, 1843, " " . + 3.3 



August 24, 1843, " " . + 1.4 



September 12, 1843, " " ... 2.2 



Hence correction to needle 2, A north +1/8, B north, 1/8 



Mean 3/6 



The above corrections have been applied. 



The preceding results indicate an annual diminution of the dip of 1.'2. To com- 

 plete the dip for the year 1843, the values for January, February, and March have 

 been inteispolated by taking the means of 1842 and 1843 of these months respectively. 

 The interpolated dips are 71 56/9, 58.'9, and 59/0. We have the final values: 



Dip for 1842.5 72 01/1 



1843.5 71 58.2 



" 1844.4 ,71 57.6 



