xxviii Introduction. 



and heavy weights, and the value of v with the N. pole of the magnet in opposite 

 directions did not differ 1° — as near a coincidence as seems attainable. 



Plane of detorsion coinciding with magnetic meridian, circle reading, 88° 16'. 



Weight attached, arms of circle turned 90°, circle reading, 358° 16'. 



Scale made to read, 150-8''". 



Magnet inserted, the scale read 150'"\ 



When the torsion circle was made to read, 290° 0'. 



V = 6S" 16', k = 0-0001303. 



Oct. 20*^. The general readings of the scale having so increased as to render 

 the middle of the scale (150) no longer the mean position, the arms of the torsion 

 circle were turned 50'. Circle reading 289° 10'. 



V = 69° 6', ^' = 0-000 L248. 



31. The time of each vibration of the bifilar magnet had been determined by 

 Mr Russell to be about 25 seconds ; his observations are not recorded. 

 The following observations were made. 



June 25^ 8\ 



Mean of 7 « 

 4 



estimations, 



2600 

 26-32 



5 



. . . 



26-63 



4 



. . . 



27-03 



7 



... 



26-40 



July W Q" 



Mean of all, 1 vibration =26^.48. 



Mean of 8 vibrations, 26-25 

 6 ... 2617 



8 ... 26-18 



Mean of all, 1 vibration = 26.20. 



The above vibrations were obtained by means of an iron key. Observations 

 when the magnet was vibrating naturally gave a result of 25 seconds. 



July 18"^ 20^". Observations during a natural vibration : 



s. 



14 vibrations, 35*52 1 vibration, 25-37 



20 ••• 53-10 1 ••• 26-55 



In each of these only two estimations were made, one at the commencement, 

 and one at the termination. The last set is not considered good, as the vibration 

 had nearly ceased when the last estimation was made. 



