7. But it happened not infrequently that plates, after being heated,
showed no power of combining oxygen and hydrogen gases, though left
undisturbed in them for two hours. Sometimes also it would happen that a
plate which, having been heated to dull redness, acted feebly, upon being
heated to whiteness ceased to act; and at other times a plate which, having
been slightly heated, did not act, was rendered active by a more powerful
ignition.

598. Though thus uncertain in its action, and though often diminishing the
power given to the plates at the positive pole of the pile (584.), still it
is evident that heat can render platina active, which before was inert
(595.). The cause of its occasional failure appears to be due to the
surface of the metal becoming soiled, either from something previously
adhering to it, which is made to adhere more closely by the action of the
heat, or from matter communicated from the flame of the lamp, or from the
air itself. It often happens that a polished plate of platina, when heated
by the spirit-lamp and a blowpipe, becomes dulled and clouded on its
surface by something either formed or deposited there; and this, and much
less than this, is sufficient to prevent it from exhibiting the curious
power now under consideration (634. 636.). Platina also has been said to
combine with carbon; and it is not at all unlikely that in processes of
heating, where carbon or its compounds are present, a film of such a
compound may be thus formed, and thus prevent the exhibition of the
properties belonging to _pure_ platina[A].

[A] When heat does confer the property it is only by the destruction
or dissipation of organic or other matter which had previously soiled
the plate (632. 633. 634.).--_Dec. 1838._

599. The action of alkalies and acids in giving platina this property was
now experimentally examined. Platina plates (569.) having no action on
mixed oxygen and hydrogen, being boiled in a solution of caustic potassa,
washed, and then put into the gases, were found occasionally to act pretty
well, but at other times to fail. In the latter case I concluded that the
impurity upon the surface of the platina was of a nature not to be removed
by the mere solvent action of the alkali, for when the plates were rubbed
with a little emery, and the same solution of alkali (592.), they became
active.

600. The action of acids was far more constant and satisfactory. A platina
plate was boiled in dilute