CONTRIBUTIONS TO SCIENCE. 523 



to show its movements. On exciting the magnet the liquid 

 rotated in one direction, and on reversing the polarity of the 

 magnet the direction of rotation was reversed. If the plates 

 be suspended by a string, so that they can readily turn round 

 in the beaker about a vertical axis, the action of the magnet 

 on the current in the vertical wire will cause the plates to 

 turn always in the direction opposite to that of the liquid. 



The laws of the mechanical action of conductors convey- 

 ing currents upon magnets and upon each other were in- 

 vestigated by Ampere in a series of experiments which were 

 at once conclusive and exhaustive,. These experiments were 

 alluded to in the highest terms by Professor Maxwell. Any 

 account of them would be out of place here, and we only 

 refer to them as furnishing the experimental evidence for the 

 statements which follow. 



We have already described the manner in which magnetic 

 lines of force may be supposed to surround a wire conveying 

 a current. Now let such a wire be bent into a closed curve 

 or ring which need not necessarily be circular. The lines of 

 force, which themselves form closed curves around the wire, 

 will all pass in the same direction through the ring formed 

 by the wire conveying the current, as if they were strung 

 upon the wire, and hence the North pole of a magnet will 

 tend to pass through the ring in the direction of the lines of 

 force ; and a moment's reflection will show that this direction 

 is that in which a right-handed screw would advance if 

 rotating in the direction of the current in the wire. Hence, 

 if the North pole of a magnet be brought near to such a 

 small closed circuit, on the one side it will be attracted and 

 tend to pass through the circuit ; on the other side it will be 

 repelled. The South pole of a magnet will be acted upon in 

 precisely the opposite manner. Hence if a small magnetic 

 needle be suspended within a coil of wire conveying -a cur- 

 rent, it will tend to set itself at right angles to the plane of 

 the coil. Such an arrangement constitutes a galvanometer. 



Now suppose that we have a small disc of steel of the 

 same size and shape as the ring formed by the wire, and 

 that this disc is magnetised so that one side is a north pole 



