{SED 5) 
hydrogen temperatures, or as KORNIGSBERGER — In a manner leading 
to a similar dependence upon temperature — explains the pheno- 
menon that was then supposed to exist, by the recombination of the 
electrons that had been freed by dissociation. I already questioned 
the validity of this assumption with respect to its application to 
perfectly pure metals at hydrogen temperatures, when the latest 
experimental results (Comm. by KAMERLINGH Onnes and Cray) obtained 
with extremely pure gold showed that the point of proportionality 
would always have to be sought at still lower temperatures. It is 
now quite clear that in the case of metals like gold and platinum 
at any rate that theory must be dropped. It seems that the 
free electrons in the main remain free, and it seems to be the 
movable parts of the vibrators that are now bound, their motion 
at ordinary temperature forming the obstacles to conduction ; these 
disappear when the temperature is lowered sufficiently as the vibrators 
become then practically immovable *). There is, in the meantime, no 
occasion to caleulate, unless for still much lower temperatures which 
cannot just yet be realised, a “latent heat of vaporization” or a 
“dissociation constant’ for the electrons for the case of pure metals 
of the type treated. 
The marked decrease in the resistance until it becomes practically 
zero at a temperature just above 4° Kk. and its remaining at this 
value as the temperature is lowered further as has been shown over 
a range of about two and a half degrees, so that, as far as resistance 
of these metals is concerned, the boiling point of helium is practically 
the absolute zero, points in another direction. It seems to me to be 
connected with the change with temperature of the heat energy of 
molecular motion of solid substances that has been deduced by EINSTEIN 
in his theory of the specifie heats, on the assumption that it is the 
energy of vibrators determined by radiation equilibrium. 
In particular an obvious assumption to make is that the mean 
free path of the electrons which provide conduction is determined 
by the elongation of the above mentioned vibrators. To further 
illustrate this point let us keep as closely as possible to the theory 
of electrical resistance of Rircke ®, Drupe and Lorentz, who has 
developed it into a pure theory of electrons. We take the formula 
1) That the vibrators become practically immovable represents what we have 
formerly called the “freezing” of the electrons. 
2) Rrecxe, Physik. ZS. 1909, p. 512 
