46 EMISSION OF ELECTRICITY FROM HOT BODIES

evidence favouring the intrinsic theory has accumulated. Thus
Richardson and Comptonl examined the photoelectric currents
obtained when monochromatic light fell on small discs of vari-
ous metals placed at the centre of a large spherical electrode,
With this arrangement the saturation value of the current
should be reached when there is no difference of potential be-
tween the two electrodes* This was found to be the case if
the contact potentials were included among the potential dif-
ferences operative. Somewhat similar experiments have been
made by Page.2 In all these experiments good vacuum
conditions were attained. In addition, in Richardson and
Compton's experiments with sodium, and in all Page's experi-
ments, the metal surfaces tested were cut mechanically in
<vacuo. Still more recently the contact difference of potential
has been measured directly under the best vacuum conditions
with surfaces machined in vacua by A, E, Hennings,1 who
finds that the potential differences are still of the order usually
observed, the metals being more electropositive when freshly
cut All these experiments support the intrinsic potential
theory, although there is abundant evidence that pro-

duce definite and complicated changes in the observed values*
On the other hand, Hughes,4 working with surfaces of metals
freshly distilled in vacuo^ found the metals to be initially most
electronegative and to become more electropositive under the
action of small quantities of air, Millikan and Bonder/ also,
have found that surfaces of sodium are most electronegative
when freshly cut and become.more electropositive on oxida-
tion. It is clear that the experimental evidence as to the
origin of contact potential differences is still conflicting, but
the balance would seem to favour the view that it is due to
an intrinsic property of the materials and • not to surface films
of foreign matter,

. *•" Phil .Mag.," VoL XXIV, p. 575 (1913) • cf. also K* T. Compton, ibid.,
VoL XXIII, p, 579 (*9«).

* " Amw. Jour, Scl," VoL XXXVI, p, 501 (1913),
8 " Phys, Rev,,'1 VoL IV, p, aaS (1914).
*" Phil Mag./' Vol. XXVHI, p, 337' Quite recently a considerable amount of