284 Mr. Faraday on an 



bends downwards, and is continued till it just dips into the 

 centre of the mercurial surface. The wire is diminished in size 

 for a short distance above the surface of the mercury, and its 

 lower extremity amalgamated, for the purpose of ensuring good 

 contact ; and so also is the copper pin at the bottom of the cup. 

 When the poles of a voltaic apparatus are connected with the 

 brass pillar, and with the lateral copper wire, the upper pole of 

 the magnet immediately rotates round the wire which dips into 

 the mercury ; and in one direction or the other, according as the 

 connexions are made. 



The other vessel is of the form delineated in the plate. The 

 stem is hollow and tubular ; but, instead of being filled by a plug, 

 as is the aperture in the other vessel, a small copper socket is 

 placed in it, and retained there by being fastened to a circular 

 plate below, which is cemented to the glass foot, so that no mer- 

 cury shall pass out by It. This plate is tinned and amalgamated 

 on its lower surface, and stands on another plate and wire, just 

 as in the former instance. A small circular bar magnet is placed 

 in the socket, at any convenient height, and then mercury poured 

 in until it rise^ so high that nothing but the projecting pole of the 

 magnet rs left above its surface at the centre. The forms and rela- 

 tive positions of the magnet, socket, plate, &c. are' seen in fig. 2, 



The cross wire supported by the brass pillar is also prolonged 

 on the right hand, until over the centre of the vessel just de- 

 scribed ; it then turns downwards, and descends about half an 

 inch : it has its lower extremity hollowed out into a cup, the 

 inner surface of which is well amalgamated. A smaller piece 

 of copper wire has a spherical head fixed on to it, of such 

 a size that it may play in the cup in the manner of a ball and 

 socket-joint, and being well amalgamated, it, when in the cup, 

 retains sufficient fluid mercury by capillary attraction to form 

 an excellent contact with freedom of motion. The ball is pre- 

 vented from falling out of the socket by a piece' of fine thread, 

 which, being fastened to it at the top, passes through a small 

 hole at the summit of the cup, and is made fast on the outside 

 of the thick wire. This is more minutely explained by Fig. 3, 

 and 4. The small wire is of such a length that it may dip a 



