1822.] Historical Sket oh of Electro-ma gnetis7)i. 109 



vM. Berthollet published in the Annales de Chimie, xvi. 113, and 

 are accompanied by some theoretical notions very different to 

 those of M. Oersted. Instead of using a round wire to con- 

 nect the two poles of the battery, M. Berzelius employed 

 bands of tin, and parallelopipeds, and concludes that they present 

 the magnetic phenomena under better circumstances for obser- 

 .wvation than the round wire. His conclusion is, that the internal 

 magnetic state of a transverse section of the wire 

 may be represented by two magnets placed with 

 their opposite poles together as in the figure ; so 

 that if the wire used be square, it will then be a 

 metallic parallelopiped through which the electric 

 current moves, each of its angles will be a mag- 

 netic pole, equal in extent to the length of the 

 parallelopiped, through which the current is passing: the opposite 

 angles will be magnetic poles of the same kind; while those 

 which terminate the same face will have different poles. Hence 

 in passing a needle round the wire, four poles should be found, a 

 north, a south, a north, and a south .^ 



M. Berzelius remarks also, that it appears each electricity (for 

 he supposes two) is represented in the wire by its own magnet, 

 and that each has its analogous magnetic pole turned to the 

 same side as regards its direction. It is evident, he says, that 

 the ordinary magnetic phenomena differ from those of a current 

 in this, that in the latter case there is a double and inverse pola- 

 rity, while in common magnets there is only simple polarity, and 

 though the double magnetic polarity may be readily imitated 

 artificially, there are no means known of imitating by electricity, 

 the simple magnetic polarity. 



M. Berzelius thinks that this exposition explains all the pheno- 

 mena that have yet been observed, and will explain all those 

 that shall be ; for, he says, it is sufficient to foretell all those 

 of which the conducting body is in this state susceptible. He 

 considers M. Ampere as quite wrong in his conjectures, and the 

 hypotheses of M. Oersted, though ingenious, yet very impro- 

 bable. It is, however, probable that M. Berzelius has been 

 himself too hasty in his conclusions. The state of the wire, indi- 

 cated in the section above, is utterly incompatible with the expe- 

 riments of M. Oersted and others, as may readily be seen by 

 reference to fig. 2, t), 7, 8, 9, 11, of PL IX, vol. ii. Annals, New 

 Series, and to the phenomena they are intended to illustrate, 

 indeed it is only necessary to experiment with a view of ascer- 

 taining the four supposed poles in the angles of a square wire, 

 and it will be immediately found that instead of any particular 

 angle exhibiting a constant polarity, it will present the pheno- 

 jm^na of either a north or south pole as the needle approaches 

 dram the one or the other side towards it. There can, how- 

 e¥er, be but little doubt that M. Berzelius will correct his opi- 



* Annals of Philosophy^ ii. 287, New Series, " 



