Abstract
It is shown that an expansion of the electronic part of the conductivity of an amorphous material in integral powers of the potential is not valid unless an infinite number of terms of the expansion are kept. These terms are summed and a new expansion is made in integral powers of a restricted class of potentials for which it is possible to keep only a finite number of terms. It is suggested that this is the solution to several difficulties which have arisen in the field in the past few years. Specific attention is drawn to the case of liquid Na.

This publication has 11 references indexed in Scilit: