lder diminished the current in the same high proportions.

[A] Bibliotheque Universelle, 1837, vii. 388.

1638. That one electrode should have this striking relation to heat whilst
the other remained absolutely without, seem to me as incompatible with what
I conceived to be the character of a current as unipolarity (1627. 1635.),
and it was therefore with some anxiety that I repeated the experiment. The
electrodes which I used were platina; the electrolyte, water containing
about one sixth of sulphuric acid by weight: the voltaic battery consisted
of two pairs of amalgamated zinc and platina plates in dilute sulphuric
acid, and the galvanometer in the circuit was one with two needles, and
gave when the arrangement was complete a deflexion of 10 deg. or 12 deg..

1639. Under these circumstances heating either electrode increased the
current; heating both produced still more effect. When both were heated, if
either were cooled, the effect on the current fell in proportion. The
proportion of effect due to heating this or that electrode varied, but on
the whole heating the negative seemed to favour the passage of the current
somewhat more than heating the positive. Whether the application of heat
were by a flame applied underneath, or one directed by a blowpipe from
above, or by a hot iron or coal, the effect was the same.

1640. Having thus removed the difficulty out of the way of my views
regarding a current, I did not pursue this curious experiment further. It
is probable, that the difference between my results and those of M. de la
Rive may depend upon the relative values of the currents used; for I
employed only a weak one resulting from two pairs of plates two inches long
and half an inch wide, whilst M. de la Rive used four pairs of plates of
sixteen square inches in surface.

* * * * *

1641. Electric discharges in the atmosphere in the form of balls of fire
have occasionally been described. Such phenomena appear to me to be
incompatible with all that we know of electricity and its modes of
discharge. As _time_ is an element in the effect (1418. 1436.) it is
possible perhaps that an electric discharge might really pass as a ball
from place to place; but as every thing shows that its velocity must be
almost infinite, and the time of its duration exceedingly small, it is
impossible that the eye should perceive it as anything else than a line of
light. That phenomena of balls of fir