592 



MAGNETIC FIELD. 



The readings are made by a very sensitive method, described 

 above (661) ; by reading with the lenses /, when the point of the 

 needle is made to coincide with its real image in a concave mirror M 

 supported by a vertical circle. 



1164. The sources of error are here more numerous than in 

 the declination compass. For in order that the inclination could 

 be determined by a simple reading, it is necessary : ist, to know the 

 horizontal line on the vertical circle ; 2nd, that the needle be centred 



Fig. 242. 



in the circle ; 3rd, that the magnetic axis of the needle be parallel to 

 the line of sighting ; 4th, that the centre of gravity of the needle be 

 on the axis of rotation ; 5th, and lastly, that the plane of the agates 

 be horizontal. All these conditions can only be approximately 

 realised. 



i st. In order to know the horizontal line or the vertical, the 

 needle is replaced by a plumb-line situate in the same plane, and 

 this wire is observed successively above and below in the two 

 mirrors; the mean of the readings gives the division which is at 



