63 



It should be observed that the two observations for the 

 determination of m may be made in a room, where the mag- 

 nets are under the action of local disturbing forces ; it is 

 only necessary that these forces should not be so great as 

 to alter the magnetic distribution in the deflecting bar, and 

 that they should remain unchanged during the observation. 

 This circumstance, of course, will contribute to the facility 

 of the observation, and to the exactitude of the result. It 

 will, probably, 'not be necessary to repeat these observations 

 on every occasion on which the value of R is sought by de- 

 flection ; the repetition being, in fact, unnecessary so long as 

 the moment of the deflecting bar continues unchanged. 



For the observation of deflection it is only required that 

 the inclinometer should be provided with a revolving arm, 

 moveable round the centre of the divided circle, for the sup- 

 port of the deflecting magnet ; while a second arm, connected 

 with the former, and at right angles to it, carries the micro- 

 scopes by which the position of the needle is observed. The 

 general plan of the instruments, now in course of preparation 

 for the Arctic expeditions, is similar to that of one made 

 for me by Mr. Barrow in 1846 (see Proceedings, Vol. III., 

 No. 56). The plane of the divided circle is separate from that 

 in which the needle moves, but parallel to it ; and there is an 

 adjustment, by which the axle of the needle is brought to 

 coincide in direction with the axis of the divided circle. The 

 circle is six inches in diameter ; it is divided to 10', and read, 

 by verniers, to one minute. The numbering of the gradua- 

 tion commences at each extremity of the horizontal diameter, 

 and extends to 180°. The needle is three inches and a half 

 long; and is enclosed (together with its supports) in a rectan- 

 gular wooden box with glazed sides. The microscopes by 

 which its position is observed carry each a line in the focus, in 

 the direction of the radius of the circle ; and the position of 

 these lines is adjusted by the same means as those employed 

 in the former adjustment. 



