BiFiLAR OR Horizontal Force Magnetometer. xxxi 



39. The value of the coefficient deduced from the angle of torsion of the sus- 

 pending wire is 



1847. k = acotv = 0-00032675 x cot 69°-3' = 00001251. 



135 



The ratio of the true value of k to that determined by the angle of torsion = = 108. 



125 



40. The true values of k from 1841, obtained from the formula 1-08 x a cot v 

 are given below : — 



July ll-i 201^ 1841— July 23'^ 5^1841, A; = 0000128 



Aug. 4 20 1841— Sept. 7 5 1841, A; = 0-000164 



Sept. 7 20 1841— Sept. 30 5 1841, A;=0000158 



Oct. 6 20 1841— Oct. 19 28 1841, A; = 0-000141 



Oct. 19 23 1841— April 27 4 1843, A;=0 000135 



April 28 2 1843— Nov. 8 22 1843, . . . . • A;=0-000130 



Nov. 10 8 1843— Dec. 31 12 1845, A; = 0000140 



Jan. 12 1846, and afterwards, A = 0-000135 



In order to reduce the variations of the horizontal component given, pages 229 

 to 238, in the volume for 1843, to their true values in parts of the whole horizontal 

 component, they must be multiplied by the factor 1-316. 



41. The bifilar magnet was adjusted November 10"^ 1843, when the angle v was 

 found = 68° 18', the bifilar scale reading 173. The angle v remained unaltered, 

 excepting for short periods during disturbances, tillJanuary 1, 1846. (See No. 43.) 



42. During considerable disturbances the collimator scale, which contains too 

 small an angle, goes out of the field of the reading telescope, it was found necessary 

 in these cases to turn the arms of the torsion circle until it reappeared ; afterwards 

 the arms of the torsion circle were turned to their original position : experiments 

 were made in the end of 1842, during periods of slight change, which shewed, after 

 turning the arms of the torsion circle a few degrees in either direction, that on re- 

 curring to the original value of v, the scale readings were unaltered. If j3 be the 

 small angle through which the arms of the torsion circle are turned, n be the scale 

 reading trvlnus 170 (the adopted scale zero), then N, the number of scale divisions 

 from the zero (corrected for temperature) for the same force when ^ = 0, is ob- 

 tained from the formula,* 



/3 



2sin2 / B\ cos {v + ^) ^ ^ , 



- cos ( V + o I 4 w 5^ ^ + t q' 



V \ 2 / cos V 



A + w B +t ^ 



N=- 



a cos 



^ is considered negative when v is diminished, n is negative when the reading is 

 below the zero (170), t is the temperature of the magnet minus 26°, and q' is the 

 temperature coefficient in scale divisions. 



* Introduction, 1843, p. xxxiii. 



