316 OBSERVATIONS OF THE MAGNETIC DIP 



by that process the facility of measuring the dip of the needle several times 

 in each position. For the purpose of designating the magnetic poles, the two 

 ends on one side of each needle, are marked A and B. A spirit level is at- 

 tached to the uppermost normal point of the vertical circle, and is levelled by 

 three foot screws. 



The method of observation is as follows: 



1. The plane of the vertical circle being in the magnetic meridian, the 

 graduated face of the instrument and the marked side of the needle to the east, 

 I wait for the needle to come to a state of rest, and then read off its position at 

 its two extremities for the purpose of correcting the eccentricity of the needle. 

 Should the observation be unexpectedly disturbed a little, by a brisk or strong 

 wind, as is sometimes the case, then I measure the position of the needle 

 when the arc of vibration is reduced to about ten or fifteen minutes, taking 

 the mean of the extreme oscillations at both ends. 



Without disturbing the position of the instrument, I now, by the aid of the 

 brass Y's, lift up the needle and gently let it down to bear again on its agate 

 supports. When at rest, I read off again as before. I repeat the same opera- 

 tion three times, and each time take the mean of both ends; thus obtaining 

 three mean readings, the sum of which being divided by three, gives the mean 

 of six readings in the same position of the instrument and needle. The three 

 mean readings generally differ but by very few minutes; the extreme differ- 

 ence, so far, with my instrument, very seldom goes to ten minutes. 



2. I correct the want of parallelism of the zero line and level, by turning 

 slowly the vertical circle 180° azimuth, which brings the face of the instru- 

 ment and the marked side of the needle from the east to the west, and there I 

 make six readings, the mean of which taken as before, gives the dip in this 

 second position of the instrument and needle. 



Now, was the needle of a perfect construction, the mean of the two results 

 obtained in the two preceding positions would be the true magnetic dip at the 

 place of observation. But, as such perfection cannot be expected, it becomes 

 necessary to recur to the following operations for the purpose of compensating 

 all the errors arising from the position of the magnetic axis of the needle. 



3. Leaving the instrument in its second position, I proceed to correct 

 the want of coincidence between the magnetic axis, and the axis of figure of 

 the needle, by turning in the brass Y's, the marked side of the needle from 

 the west to the east, and repeating the six readings as above stated. 



