1821.] Historical Sketeh of Electro-magnetism. 287 



bringing the two poles together, its action on a magnetic needle 

 was much diminished. 



In a letter (without date) from M. Berzelius to M. Berthollet r 

 published in the Annates de Chimie, for Feb. p. 113, an expe- 

 riment is described which consisted in placing a thin leaf of tin r 

 eight inches long and two inches wide, parallel to, and in the 

 plane of, the meridian, and in that position connecting it with 

 the elements of a voltaic circle. A magnetic needle brought 

 near the lower edge of this plate was thrown 20° from the mag- 

 netic meridian. On moving it slowly upwards, it took its natural 

 position, when level with the middle of the plate, except that it 

 was raised at one end, and depressed at the other ; and when 

 near the upper edge, it moved 20° from the magnetic meridian 

 in the opposite direction to what it did below. When the needle 

 was moved up and down on the opposite side of the plate, the 

 same deviation and effects took place, but in opposite directions. 

 A small portion of the upper edge of the leaf was cut, and turned 

 upwards, forming a projection above the edge. The needle 

 brought within equal distance of this projection, and the edge,, 

 was more affected by the former than the latter. 



Then using a square plate of tin, and forming the connexion at 

 opposite angles, it was found on examination that the intervening 

 angles acted more powerfully on the needle than any other parts 

 — a circumstance which proves, M. Berzelius says, that the mag- 

 netic polarity of the current goes to opposite extremities, as hap- 

 pens with electric polarity, and in artificial magnets. 



The tin band or leaf placed in a horizontal plane, and in the 

 magnetic meridian, acted on the needle just as a wire would have 

 done. The greatest deviation of the needle was immediately 

 under or above the middle of the leaf, and the edges acted as in 

 the former position. The positions assumed by the needles in 

 these experiments is exactly what would be expected. The 

 experiments receive all their interest from the way in which their 

 maker applies them to support his particular opinion, and apart 

 from that have not much new in them. M. Berzelius thinks 

 that a round wire, when made the conductor, presents a more 

 complicated case than when a square one, or a parallelopiped, is 

 used. I shall endeavour to return, however, to the theory 

 advanced by this philosopher presently. 



M. Lehoi's experiments (Feb. 27) are repetitions of what had 

 been previously made, and are only intended to prove that a 

 straight conductor can communicate magnetism in opposition to 

 the opinion of the Marquis Ridolfi, who considered that it must 

 It.' more or less turned round the needle to be magnetized. 



M. Scbweiger's experiments have nothing in them new after 

 what has been said. I do not know their date, but they are 

 inserted in the JJib. Univ. for March, 1821. The author made 

 his connecting wire pass several times round the needle, produc- 

 ing in fact tlie same accumulation of effect as in the helix, and 



