165 



is within, and the stem (bent at right angles) on the outside 

 front. 



It is intended to employ this apparatus for three pur- 

 poses ; namely, to determine, first, the magnetic declina- 

 tion; secondly, the periodical and the irregular variations 

 of the declination; thirdly, the corresponding variations 

 of the horizontal magnetic force. 



In using the instrument in the determination of the declina- 

 tion, it is to be combined with the theodolite and the transit in- 

 strument. The transit instrument is to be fixed close to the 

 southern window of the observatory; there being also an 

 aperture in the roof for the purpose of adjusting it to the 

 meridian by means of the pole star and 8 ursce minoris. The 

 centre of the theodolite is placed, as nearly as possible, at the 

 point where the line of collimation of the transit instrument 

 intersects that of the magnetic bar. "When an observation 

 is made, the telescope of the theodolite is directed to the 

 lens of the magnetic collimator, and the vertical wire of the 

 latter is made to bisect the cross in the focus of the telescope. 

 "When this is done, the line of collimation of the telescope 

 is parallel to that of the magnetic bar. But as the latter 

 line may not coincide with the magnetic axis of the bar, a 

 similar observation is to be made with the bar inverted ; and 

 the mean of the two readings will obviously give the direc- 

 tion of the magnetic meridian, freed from the error of colli- 

 mation. To determine the angle between this and the true 

 meridian, the transit telescope is to be turned over, and em- 

 ployed as a collimator. The telescope of the theodolite 

 being directed to its object glass, the middle wire in the focus 

 of the transit is to be observed, in the same manner as the wire 

 of the collimator bar was in the former part of the observa- 

 tion. The line of collimation of the theodolite telescope is 

 then in the true meridian ; and the angle read off on the 

 limb is the supplement of the declination. 



In observing the diurnal and irregular variations of the 



