ELECTRO-MAGNETISM. 



three grains, which he "wore in a ring, 

 and which is said to have raised 746 

 grains, or 250 times its own weight, and 

 this is the greatest relative strength of 

 any magnet yet recorded. It is evident, 

 therefore, that a much greater degree of 

 magnetism can be developed in soft iron, 

 by a galvanic current, than in steel of 

 the same dimensions, by the ordinary 

 processes of magnetizing *. 



Mr. Watkins informs us that, in order 

 to obtain magnets of any power by the 

 above described method, great care 

 must be taken to ensure the purity of 

 the iron employed to form the horse- 

 shoe magnet ; and after it has been 

 welded and reduced to the proper shape, 

 it is advisable, in order thoroughly to 

 destroy any magnetism it may have ac- 

 cidentally acquired and retained during 

 the process, to heat it in a furnace, and 

 afterwards cool it very gradually, by 

 allowing it to remain undisturbed till 

 the furnace itself has grown cold. But, 

 even after every precaution has been 

 taken to ensure success, we are still lia- 

 ble to be baffled by causes which we 

 tannot explain, and which, when all cir- 

 cumstances seem to be the same, pro- 

 duce great and unexpected variations 

 in the results. It should be borne in 

 mind, indeed, that similar embarrass- 

 ments are often experienced in conduct- 

 ing almost every other experiment in 

 electro-magnetism, their results appear- 

 ing to be more or less capricious in 

 proportion as the conditions necessary 

 to be fulfilled, before uniformity can be 

 obtained, are numerous and delicate. 



(163.) The best form of a conducting- 

 \vire for exhibiting its attraction for iron 

 filings is that of a flat spiral coil, similar 

 to what is represented in fig. 66, which, 

 however, for this purpose need not be 

 rendered moveable. A wire of this 

 form, through which an electric current 

 is made to circulate, will collect a pro- 

 digious quantity of iron filings, and 

 their relative positions and arrange- 

 ment, while they remain attached, pre- 

 sent many singular appearances. If the 

 rings of wire are not continued quite to 

 the centre, but leave an opening there, 

 the particles of iron are observed to 

 arrange themselves in lines, passing 

 through the ring parallel to the axis, and 

 then closing up as radii round the edge. 

 The particles of iron in the centre erect 

 themselves into a perpendicular fila- 

 ment, in the direction of the axis of the 



Silliimin's Joutnal, quoted in the Journal of the 

 Royal Institution; 1, 609; and II. 182. 



spiral, while the intervening particles 

 form filaments inclining from the centre 

 in proportion to their distance from it*. 

 The reason of this will be evident from 

 the principle explained in $ 50. 



(164.) There appears to be a very 

 essential difference in the effects of the 

 shock of an electrical battery discharged 

 through a wire, and that of a voltaic 

 battery, in communicating permanent 

 magnetism to steel bars or needles. Mr. 

 Savary has brought to light several very 

 curious particulars relating to this sub- 

 ject, which have hitherto received no 

 explanation t. When the discharge 

 from a Leyden battery is made through 

 a straight wire, different needles, though 

 equal in size, and parallel to each other, 

 and placed transversely on the same 

 side of the wire, but at different dis- 

 tances, have their polarities not disposed 

 in all of them in the same manner. In 

 some the poles have the same relative 

 situation as those of a needle previously 

 magnetized, and free to move, which 

 has taken the position it would have 

 when under the influence of a continued 

 voltaic current passing in the same di- 

 rection along the wire. 



But in others the position of the in- 

 duced poles is the reverse of this. For 

 the sake of conciseness of expression, 

 we shall call the action which produces 

 an arrangement of poles similar to that 

 resulting from a voltaic current, positive 

 magnetization ; the contrary effect being 

 that of negative magnetization. Thus, 

 in a series of experiments in which the 

 needles were placed at distances from 

 the wire which increased by equal in- 

 tervals, at the point of contact wilh the 

 wire the needle was magnetized posi- 

 tively, at a small distance negatively ; 

 a little further off it had acquired no 

 magnetism whatever ; at a distance 

 somewhat greater than this, it exhibited 

 positive magnetism ; and this effect con- 

 tinued for a certain interval, beyond 

 which the magnetization was again ne- 

 gative. When still more remote, it was 

 positive, and continued so to all greater 

 distances that were tried. Hence the 

 action appears to be periodical with re- 

 lation to the distance at which it is ex- 

 erted. 



(165.) The number of periods in these 

 alternations, as well as the distances at 

 which they occur, appear to depend upon 

 a variety of circumstances of which it 



* Watkins, Popular Sketch of Electro-Magnetism, 

 p. 46. 



| Amwles de Chimie, tome, xxxiv, 



