BIFILAR OR HorIZONTAL ForcE MAGNETOMETER. x1 
ture on the bar must be eliminated by means of temperature corrections applied 
to the observations. 
Our object will therefore be to ascertain—1°, The value of one scale division ; 
and, 2°, The total temperature correction for 1° F.; both of these in parts of force, 
that is to say, assuming the whole horizontal component of the earth’s magnetic 
force as equal to unity. 
The following were the values of these coefficients at the beginning of 1847 :— 
Value of one scale division in parts of force & = k = 0:0001385. (Intr. 1845-46, p, xxxi.); 
Total Temperature correction for 1° F. in parts of foree= g = 0:000266, (Intr, 1845-46, p. xlvi.) ; 
and consequently the temperature correction in scale divisions 
This state of things remained until a change was made in the plane of the 
vertical force magnetometer, in January 1848. 
When the magnet of this instrument was in a plane perpendicular to the 
magnetic meridian before the change, its effect upon the bifilar was ascertained on 
July 30, 1841, to be = — 3:15 scale divisions = — 0:000517 in parts of force, 
and from observations, September 4, 1843, it was found = —4:03 scale divisions= 
— 0°000524 in parts of force; the mean of the twois — 0:000520. No correction 
on this account had been applied to the bifilar readings (Introd. 1843, p. xxxiv.). 
The effect of the balance needle in its new position in the magnetic meridian 
was found on January 24, 1848, to be = — 1°58 scale divisions = — 0:000213 in 
parts of force. Consequently, in order to make the observations after January 
18, 1848, comparable with those before that date, a correction of — 0:000307 parts 
of force = — 2°3 scale divisions, must be applied; and accordingly this has been 
done to all the observations after January 18, 1848, and before July 8, 1851. 
After July 8, 1851, the value of a scale division of the bifilar became less in the 
proportion of 23 to 25 nearly; consequently, the correction on account of the 
balance magnet, which was = — 2°3 scale divisions before that date, became = —2°5 
scale divisions after it. This amount has accordingly been deducted from all the 
observations of the bifilar after July 8, 1851. 
On July 8, 1851, at 23" 54™ Gottingen mean time, a change was made upon 
the bifilar magnetometer, the torsion circle being turned. 
The readings of the torsion circle before being turned were, A 109° 31’, B 
289° 33’. After being turned they were, A 107° 59’, B 288° 0’. The mean 
amount of turning was therefore 1° 32’:5. 
Before the change 
v= (69° 375 . (Introduction, 1845-46, p. xxxii.) 
hence, after the change 
vy = 70° 36’. 
MAG, AND MET. oBs. 1847 To 1855. ¢ 
