INTRODUCTION. 



degree Fahrenheit was found to be q = 0.0003937G2. Representing the two temperatures by t 

 and t , and the observed time of a single vibration by T , the corrections applicable to T may bo 

 combined in the formula 



Dr. Lament has pointed out a further very small negative correction necessary in obtaining 

 m X, due to the change in the value of w, produced by the earth's inducing action on the de- 

 flecting magnet in its two positions. In one case the magnet is perpendicular to the magnetic 

 meridian, or very nearly so, when the magnetic moment is that proper to the bar itself; and in 

 the other, it coincides with that meridian, and its moment is augmented by the inducing action 

 of the earth. The determination of this coefficient requires the use of a bifilar magnetometer ; 

 and as there was no such instrument at command, the correction has necessarily been omitted 

 in the calculation of our observations. 



MODE OF OBSERVING, AND EXPLANATION OF THE OBSERVATIONS. 



Observations commence with experiments of vibration. The days selected were the 1st, llth, 

 and term-day of each month. When either of the days first named fell on Sunday, observa- 

 tions were made on the following day. 



The vibration box being placed on the stand about 11 A. M. of the assigned day, its tele- 

 scope and counterpoise were properly placed, the instrument was levelled and the torsion of the 

 suspending fibre were partially corrected at the same operation, by means of the brass plummet. 

 When the latter had come to rest, the 3.67-inch magnet and its mirror was suspended ; then ele- 

 vated by the sliding pin until the divisions of the scale were reflected into the telescope, and the 

 table-top was turned in azimuth until the vibrations were equal on each side of the scale zero. 

 When the latter were reduced to about 1 on each side of zero, the extreme scale divisions and 

 temperature of the magnet were recorded, and observations immediately commenced of the times 

 of vibration across the central wire of the telescope. The times of the first and each fourth sub- 

 sequent vibration to the twenty-fourth were first noted ; then every sixtieth to the three-hun- 

 dredth, and afterwards each fourth again until three hundred and twenty-four vibrations had 

 been numbered. This completed, the extreme coinciding reflected scale divisions and tempera- 

 ture were again noted ; and finally, when the magnet had become still, the ratio of the torsion 

 to the earth's directive force was ascertained. 



But whilst awaiting opportunity for the last experiment, the unifilar magnetometer was set 

 up, the gun-metal rod secured in its place, the instrument carefully levelled, and the detorsion 

 bar suspended. When the plane of detorsion coincided (approximately) with the magnetic 

 meridian, the 3-inch magnet was suspended, and after the latter became still the circle was 

 clamped and the tangent screw turned until the image of the scale zero, seen in the mirror, 

 coincided with the vertical wire of the telescope. The readings of the verniers in (this) mag- 

 netic meridian were noted, the sliding frame adjusted at the distance of one foot to the west of 

 the suspended magnet, and the 3.67-inch deflector put in place, with its north pole to the east. 

 The telescope was then turned until the oscillations of the suspended magnet on either side of 

 the scale zero were equal ; and when they had ceased, the temperature was noted and the verniers 

 again read. The north pole of the deflecting magnet was next turned to the west, the circle 

 moved to coincidence, and its verniers and the temperature were recorded as before. Then the 

 deflector was adjusted at the distance of 1.3 foot; and when the angles of deflection in both 

 positions of its north pole had been read, it was shifted to the east arm of the graduated bar, 

 and similar experiments were made, but in reversed order of the distances. 



