

of Terrestrial Magnetism. 433 



observe the oscillations of the dipping-needle itself. This method 

 has, however, several inconveniences. 



The first is, that as the time of an oscillation depends on the 

 intensity of the magnetism of the needle as well as of that of the 

 earth, a change in the one might be attributed to the other, and 

 there are no means of determining in which it has taken place. 

 Another disadvantage is, that as snch experiments require a 

 considerable amount of time and no ordinary skill, they may 

 give a great deal of trouble to the observer, and yet leave it un- 

 certain whether the magnetism of the needle and of the earth 

 have remained constant in the interval occupied by the obser- 

 vation. 



For these reasons, philosophers, following the steps of Gauss, 

 have abandoned the direct measurement of the total force, and 

 for the same reasons that of the variations of the inclination; and 

 have adopted the following method of investigation. 



Conceive the total force resolved along three rectangular co- 

 ordinate axes ; one horizontal and parallel to the magnetic meri- 

 dian, another horizontal and at right angles to the magnetic 

 meridian, the third vertical. For each of these directions there 

 is a particular apparatus for determining the magnitude and the 

 variations of the components ; and from these, the direction of 

 the resultant being known, the value and variation of the total 

 force and inclination are easily deduced. The instrument which 

 is used to measure the horizontal component is called the unifilar 

 magnetometer. Its office is to give the absolute value of the 

 horizontal force. But the description of the method pursued 

 would lead us too far from our subject*. It may suffice to 

 say, that these experiments likewise can only be made at di- 

 stant epochs, and therefore what we have already remarked of 

 vibrations of the dipping-needle is equally applicable to them, viz. 

 that they cannot be used to determine the horary vacations, or 

 those of short duration. These last are usually deduced from 

 the variations in the position of a bar placed at right angles to 

 the magnetic meridian. The apparatus used for this is called 

 the bijilar magnetometer, a name descriptive of its construction. 



It consists of a bar-magnet suspended by two parallel threads 

 in a state of equal tension, which, were a non-magnetic body sus- 



* For the theory of this observation see Gauss's memoir Intensitas, 

 &c. ; for the practice see Weber's memoirs published in Taylor's Scientific 

 Memoirs, and the instructions in the introduction to the Annvairc Mag- 

 nftique et Meteor, of Russia. The method is shortly this : the declination 

 needle is made to oscillate, first loaded with certain weights, and then with- 

 out them. By this means the magnetic moment of the needle, multiplied 

 by the intensity of the terrestrial force, is obtained ; next, a bar magnet is 

 placed at different distances from the needle, and from the deviations which 

 it produces, the value of the force divided by the magnetic moment is de- 

 duced, and from these the magnetic force may be easily obtained. 



Phil. Mag. S. 4. Vol. 9. No. 01. June 1855." 2 F 



