00000271.htm

₂6₄ EMISSION OF ELECTRICITY FROM HOT BODIES

rather by some other substance coming from the salt. The
time changes must in fact be conditioned by something the
amount of which is determined by actions occurring at the
salt and not simply by a decomposition of cadmium iodide
vapour. In one aspect this question has been definitely
settled by Kalandyk ^l who has shown that the currents in
cadmium iodide vapour under the conditions of these experi-
ments are independent of the time, provided every trace of
water is removed from the salt and from the apparatus. The
way in which water brings about the time changes usually ob-
served is unknown. Kalandyk's experiments only tell us that
there are no time changes when water is absent, they do not
offer an explanation of the changes which occur in the presence
of water or water vapour. Sheard's results point to the con-
clusion that the large negative initial emission, when it is
present, is connected with the liberation of iodine. On this
view the smaller negative emission from the distillates would
be related to the reduced iodine content of the salt, which
after distillation probably consists of a solution of an unrecog-
nized subiodide of cadmium in CdI₂. The presence of the
subiodide would reduce the equilibrium pressure of iodine in
presence of cadmium iodide vapour. The probable existence
of a subiodide of cadmium is distinctly indicated by the work
of Morse and Jones² who succeeded in isolating a body having
the composition Cd₁₂I₂₈, probably a solution of the subiodide
InCdI₂.

It is likely that the effect of water vapour is not confined
to this particular instance and that many of the time changes
observed with other salts would not occur if all traces of water
were eliminated Such a result, at any rate, would not be
surprising if the time changes are indicative of the occurrence
of chemical reactions. For it is well known that many chemi-
cal actions which proceed very energetically in presence of a
trace of water vapour are completely inhibited in its absence.
The importance of water vapour generally for these effects is
supported by the behaviour of potassium iodide, whose vapour

•' ;"••;- • ... • * Loc. cit. ' '. '• ' • ' ' -. " • ... .". -" . •

² " Araer, Chem. Jour.," Vol. XII, p. 488 (1890).