60 Professor CuMMING on the Developement 
used, the brass is negative with’ respect to both, yet, the electric 
energies of silver and brass differ so little, that, by mcreasing the 
size of the brass, it may be made to overcome the action of the 
silver, and give a positive deviation, as in the first case of the 
experiment. The action of platina is so much more energetic than 
that of brass, as not to be overcome by the increased dimensions 
of the bar, and therefore the deviation is negative. In the last case, 
where the platina was shortened to half an inch, it became heated 
throughout, and communicated heat to the silver wire in contact 
with it; consequently the effect of the platina disappeared, and 
this case became similar to the first. To have made the series 
(Table I.) given in the first part of this paper, an accurate re- 
presentation of the electro-magnetic relations of the metals, the 
bars should have been connected with the galvanoscope, by a 
substance (if there be such an one) that should merely conduct, 
without modifying the electricity*. The electric energy of copper 
is so much inferior to that of many of the metals, that, when the 
dimensions of the bars considerably exceed those of the copper 
wires, the series thus obtained may be considered as tolerably 
accurate ; but, with others, as lead and tin for instance, it is ma- 
nifestly imperfect. The most complete scale seems to be that, 
which may be formed by using bars and wires of the same di- 
mensions, i.e. by heating equal wires of the different metals in 
contact, taken two and two together. The series given in Tables 
VIII. and IX. was formed in this manner, so far as it was prac- 
ticable. The ends of the wires to be examined, were placed, 
one in each cap of the galvanoscope; the other ends were then 
connected and dipped into a capsule filled with boiling mercury ; 
* If two small rods of antimony and bismuth, properly adjusted, were soldered together 
longitudinally, they might, perhaps, at a fixed temperature, be considered as a neutral 
conductor, 
