produced by Electricity. 47 
The facility with which experiments were made with the com- 
mon Leyden battery, enabled me to ascertain several circum- 
stances which were easy to imagine, such as that a tube filled 
with sulphuric acid of a quarter of an inch in diameter, did not 
transmit sufficient electricity to render steel magnetic; that a 
needle placed transverse to the explosion through air, was less 
magnetized than when the electricity was passed through wire ; 
that steel bars exhibited no polarity (at least at their extremities) 
when the discharge was made through them as part of the cir- 
cuit, or when they were placed parallel to the discharging wire ; 
that two bars of steel fastened together, and having the dis- 
charging wire placed through their common centre of gravity, 
showed little or no signs of magnetism after the discharge till 
they were separated, when they exhibited their north and south 
' poles opposite to each other, according to the law of position. 
These experiments distinctly showed, that raagnetism was pro- 
duced whenever concentrated electricity passed through space ; 
but the precise circumstances, or law of its production, were pot 
obvious from them. When a magnet is made to act on steel 
filings, these filings arrange themselves in curves round the poles, 
but diverge in right lines; and in their adherence to each other 
form right lines, appearing as spicula. In the attraction of the 
filings round the wire in the Voltaic circuit, on the contrary, they 
form one coherent mass, which would. probably be perfectly cy- 
lindrical were it not for the influence of gravity. In first consi- 
dering the subject, it appeared to me that there must be as many 
double poles as there could be imagined points of contact round 
the wire ; but when I found the N. and S. poles of a needle uni- 
formly attracted by the same quarters of the wire, it appeared to 
me that there must be four principal poles corresponding to these 
four quarters. You, however, pointed out to me that there was 
nothing definite in the poles, and mentioned your idea, that the 
phenomena might be explained, by supposing a kind of revolu- 
tion of magnetism round the axis of the wire, depending for its 
direction upon the position of the negative and positive sides of 
the electrical apparatus. 
To gain some light upon this matter, and to ascertain correctly 
the relations of the north and south poles of steel magnetized by 
electricity to the positive and negative state, I placed short steel 
needles round a circle made on pasteboard, of about two inches 
and a half in diameter, bringing them near each other, though 
not in contact, and fastening them to the paste-board by thread, 
so that they formed the sides of a hexagon inscribed within the 
circle. A wire was fixed in the centre of this circle, so that the 
circle was parallel to the horizon, and an electric shock was 
passed through the wire, its upper part being connected with the 
positive 
