TERRESTRIAL MAGNETISM. (APP. A.} 329 



courses ; and also for one of the quadrantal 

 courses if the compass is sufficiently affected by 

 unsym metrically placed iron to show any sensible 

 amount of the " E" constituent of quadrantal error. 

 When the deflector is to be used for determining 

 the amount of an uncorrected error, according to 

 principles (3) and (4) above, the magnetic value 

 of its scale reading must be determined by 

 experiment. This is very easily done on shore, 

 by observations of its deflecting power when set by 

 its screw to different degrees of its scale. 



APPENDIX B. 



ON A NEW FORM OF AZIMUTH MIRROR. 



\_Being extract from United Service Institution Lecture, 1878 ; 

 with additions of date 1 890. ] 



AN important objection was made to me some 

 years ago by Captain Evans against the use of 

 quadrantal correctors in the Navy, that they would 

 prevent the taking of bearings by the prismatic 

 azimuth arrangement which forms part of the 

 Admiralty standard compass. The azimuth mirror 

 applied to the compass before you was designed 

 to obviate that objection. Its use even for taking 

 bearings of objects on the horizon is not interfered 

 with by the globes constituting the quadrantal 



