Chenikal Science. 419 



depressed so low that the thin moveable wire having its lower end 

 resting'on the surface of the mercury, made an angle of about 40° 

 v;ilh the horizon. Immediately the circuit through the mercury 

 was completed, this wire began to move and rotate, and continued 

 to describe a cone whilst the connexions were preserved, which, 

 though its axis was perpendicular, evidently, from the varying 

 rapidity of its motion, regarded u line parallel to the dipping 

 needle as that in which the power acted that formed it. The di- 

 rection of the motion was, as expected, the same as that given 

 by the pole of a magnet pointing to the south. If the centre 

 from which the wire hung was elevated until the inclination of. 

 the wire was equal to that of the dip, no motion took place 

 when the wire was parallel to the dip ; if the wire was not so 

 much inclined as the dip, the motion in one part of the circle ca- 

 pable of being described by the lower end was reversed ; results 

 that necessarily follow from the relation of the dip and the mov- 

 ing wire, and which may easily be extended. 



I have described the effects above as produced by the 

 north pole of the earth, assuming that pole as a centre of ac- 

 tion, acting in a line represented by the dip of the needle. This 

 has been done that the phenomena might more readily be com- 

 pared with those produced by the pole of a magnet. M. Biot 

 has shewn by calculation that the magnetic poles of the earth 

 may be considered as two points in the magnetic axis very near 

 to each other in the centre of the globe. M. Ampere has in his 

 theory advanced the opinion that the magnetism of the earth 

 is caused by electric currents moving round its axis parallel to 

 the equator. Of the consonance existing among the calcula- 

 tion, the theory and the facts, some idea may perhaps be gained 

 from what was said, page 86, on the rotation of a pole 

 through and round a wire ring. The different sides of the plane 

 which pass through the ring, there described, and which may 

 represent the equator in M. Ampere's theory, accord perfectly 

 with the hemispheres of the globe; and the relative position of 

 the supposed points of attraction and repulsion, coincide with 

 those assigned by M. Biot for the poles of the earth itself. 

 Whatever, however, may be the state and arrangement of ter- 

 restrial magnetism, the experiments I have described bear me 

 out, I think, in presuming, that in every part of the terres- 

 trial globe an electro-magnetic wire, if left to the free action of 

 terrestrial magnetism, will move in a plane (for so the small 

 part we can experiment on may be considered), perpendicular 

 to the dip of the needle, and in a direction perpendicular to 

 the current of electricity passing through it. 



Reverting now to the expectation I entertained of altering the 

 apparent weight of a wire, it was founded on the idea that the 

 wire, moving towards the north round the pole, must rise, and 

 moving towards the south must do»cen<l ; inasmuch as a phuie 



