Magnetism mid Gcdvamsin. 171 
tive electricity to the water, as the copper does of its negative 
electricity, which would produce an accumulation of negative 
electricity in the upper part of the zinc, and of positive electri- 
city in the upper part of the copper, and the communication by 
ah did not re-establish the equilibrium by presenting a free pas- 
sage to the negative electricity from z to c, and of the positive 
electricity from c to z. The wire ah^ therefore, receives the ne- 
gative electricity of the zinc, and the positive electricity of the 
copper ; whereas a wire which forms a communication between 
the two poles of a battery, receives positive electricity from the 
pole of the zinc, and negative from that of the copper. 
“ If we attend to this distinction,^’ says M. Oersted, we 
may, with a single galvanic arc, arranged as I have described, 
repeat all the experiments which I had before made with a com- 
pound galvanic apparatus. One great advantage of this plan is, 
that we may form the arc sufficiently light to be suspended by 
a small metallic wire, so as to revolve round the axis of the wire 
prolonged ; and in this way we may examine the action of a 
magnet on the galvanic arc. 
“ For this purpose I employed the arrangement in Fig. 7. 
which is a perpendicular section of it in the direction of its breadth, 
cccc being a trough of copper 3 inches high, 4 inches long, 
and I inch wide zz^ sl plate of zinc, kept in its place by two 
pieces of cork c a brass wire about a quarter of a 
line in diameter ah^ b. brass wire as small as possible, but ca- 
pable of sustaining the apparatus ; and cac, a linen thread for 
attaching the wire to the apparatus. The trough contains the 
usual conducting fluid. The uniting wire of this apparatus will 
attract the north pole of the needle when it is placed on the left 
side of the plane cffffz^ regarded in the direction fz. On 
the same side, the south pole will be repelled. On the other 
side of this plane, the north will be repelled and the south pole 
attracted. In effecting this, we must not place the needle above 
nor below or jfc. If, instead of presenting a small move- 
able needle to the uniting wire, we present to one of the extre- 
mities one of the foles of a strong magnet, the attraction or 
* These dimensions may vary to infinity, only the breadth should not be great, 
and the trough made of as thin plates as possible. 
