Mr Barlow 07i the Laws qf^ Electro-Magnetic Action. 107 
observations were made^ were situated as shewn at B, C 5 D. 
At P and N were placed two cups of mercuryj and from the 
battery K proceeded two stout conducting wires, which were 
immersed in the cups at P and N ; and, by immersing also the 
extremities of the wires at P and N in the same cups, the cir- 
cuit was carried through the whole wire. We thus began our 
observations on the whole length of 838 feet ; then unwound 
two circumferences from each frame, thereby shortening the 
circuit 40 feet, or reducing it to 798 feet ; we then unwound 
40 feet more, and so on to the end, as detailed in the following- 
tabulated results. 
There still, however, remained to guard against the variable 
power of the battery while the experiments were in progress, which 
occupied several hours. This was as follows : Between N and P 
was situated another compassA, over which could be placed a short 
conducting wire, which, prior to each experiment, was made to 
unite the two cups N and P, and its effect on the compass was re- 
gistered; immediately after which, the wires P and N were immer- 
sed, and the short conductor removed ; by which means the rela- 
tive power of the battery became known at each observation ; for 
it is shewn in Essay on Magnetic Attractions,” &c. (art. 247.), 
that, all other things being the same, while the wire is placed in 
the magnetic meridian the power is directly as the tangent of 
the needle’s deviation, or rather the latter is as the former. It is 
thus easy to compute what the several deviations would have 
been, had the power of the battery remained constant, as is done 
in Table II. It may be proper to state, that the apparatus em- 
ployed was Dr Hare’s calorimeter, and that it was raised out of 
the acid after each set of experiments ; that is, as soon as the 
battery was immersed, the short conductor was placed in the cups, 
and the deviation on the compass A registered. I’his was now 
removed, the wires P and N inserted, and the deviations on the 
three compasses B, C, D, were registered by three observers, 
one at each. The distance of the conducting wire from the 
needle was only half an inch. These wires P, N were now removed, 
the battery remaining down, and the short wire again inserted, 
the compass A registered, and the observation on the other three 
compasses taken as before; thus obtaining two sets of observations 
at each immersion, corresponding to the distance of half an inch. 
Then, in the latter part of the series, the battery being still 
