44 REPORTON 



would be necessary that it should be moving in the opposite direction. 5. The 

 extreme scale readings attained by the magnet at the eastern and western 

 extremities of its arc of vibration were again observed; and then the thermometer 

 in the magnet-box was read. 6. The necessary observations for determining the 

 coefficient of torsion of the suspension fibres were made. When the instrument was 

 properly adjusted for observation the torsion circle always read 300. With it 

 remaining at that reading the arc of vibration of the magnet was reduced to four 

 or five scale divisions (by means of the magnetized screw-driver) and then the scale 

 was read. Next the torsion circle was turned backward one-quarter of a revolution, 

 so as to make it indicate 210, and the scale was again read. After that.the torsion 

 circle was turned forward half a revolution (passing through the point 300), so as 

 to make it indicate 30, and the scale was read. Finally, the torsion circle was 

 turned backward one-quarter of a revolution, so as to make it indicate 300, and 

 the scale was once more read. Subtracting the second scale reading from the first, 

 the second from the third, and the fourth from the third, gave three differences, 

 which were added together and divided by four. The result was the number of 

 scale divisions through which the magnet was deflected by a twist of ninety degrees 

 in the suspension fibres. 



Observations of Deflections were made as follows : The instruments having been 

 set up and adjusted in the manner already explained, the short magnet, S. 8, was 

 suspended in the magnet-box, and the telescope having been pointed so that its 

 vertical wire cut the magnet scale approximately at its central division (not neces- 

 sarily the magnetic axis) the horizontal limb of the theodolite was clamped firmly. 

 Then, 1. The time was noted. 2. The thermometer inside the magnet-box 

 was read. 3. The long magnet C. 32 (which we will now call the deflecting 

 magnet) was placed on the deflecting bar support, with its axis east and west, its 

 centre on a level with and at a distance of two feet to the west of the suspended 

 magnet, and its north end west; the vibrations of the suspended magnet were 

 reduced to four or five scale divisions, by means of the magnetised screw-driver, 

 and then its scale was read. 4. The deflecting magnet (remaining in the same 

 place on the deflecting bar support as before) was reversed end for end, so as to 

 bring its north end east, and the scale of the suspended magnet was read. 5. The 

 reversals were repeated twice more, so as to give in all two scale readings with the 

 north end of the deflecting magnet to the west, and two scale readings with it to 

 the east. The mean of the two scale readings obtained with the north end of the 

 deflecting magnet west, were subtracted from the mean of the two scale readings 

 obtained with its north end east. The difference was twice the value of the angle 

 of deflection, as resulting from observations made with the deflecting magnet west 

 of the suspended magnet. 6. The deflecting magnet was lifted from the deflecting 

 bar support to the west, and placed on that to the east, of the suspended magnet; 

 its distance from the suspended magnet being still two feet, and its north end being 

 to the east, the scale of the suspended magnet, was read. 7. The deflecting mag- 

 net (remaining in the same place on the eastern deflecting bar support) was reversed 

 end for end, so as to bring its north end west, and the scale of the suspended mag- 

 net was read, 8, The reversals were repeated twice more, so to give in all two 



