or the - at the right or the left. The plates are made of
four different sizes: No. 1, 19 of which serve for an accumulator of 1
square meter; No. 2, 21, 25, or 29 of which serve for accumulators of 2, 3,
and 4 square meters; No. 3, which with 21, 25, or 29 plates composes
accumulators of 5, 6, and 7 square meters; and No. 4, which with 21, 23,
25, 27 or 29 plates forms accumulators of 8, 9, 10, 11, and 12 square
meters.

As the plates are entirely submerged in the liquid their entire surface is
active, and the entire surface being absolutely flat, it is sufficient to
preserve their respective distance at any one point in order to have it
everywhere alike. The weight of the plate depends on the intended duration
of the plate and its capacity. As for the negative plate, its thickness is
the most important factor of its capacity. The proportion has yet to be
established for daily practice. The inventor uses in practice positive
plates of 0.002 meter in thickness. On the other hand, the negative plates
have a body of only 0.001 meter in thickness, their greater thickness being
due only to the deposit of compressed lead. The rod which fixes the plate
to each pole (Fig. 2) is formed of a special alloy of lead and antimony,
not attacked by acid. This gives rigidity to the rod, and hinders it from
binding when the accumulator is taken out of its case. The copper piece
which surmounts it is fitted at its base with an iron cramp, which is fixed
in the lead, and above which is a wide furrow with two grooved parts, which
being immersed in the lead hinders the copper from slipping round under the
action of the screw. The rod is square, and is cast in a single piece.
Against one of its surfaces the ends of the connected plates press flatly
up. A square form has been selected to give more surface for soldering. The
soldering is autogenous (as in the lead chambers at vitriol works). The
soldering, as well as the entire plates, is entirely immersed in the
liquid, and to prevent any leakage an insulating varnish, perfectly proof
against the acid and the current, is laid over the rod from the part
soldered upward.

[Illustration: FIG. 2.]

If it is wished to lift the accumulator from its chest for any
verification, hooks passing between the plates seize hold of the rods, and
thanks to the rigidity of the antimony lead, they effect the removal of the
apparatus without bending the rods in the least. All the parts of the
plates must be ke