In accordance with the earlier measurements 1) 2) the resistance-temperature curves of the gold wires measured show a minimum.
The influence of the annealing time remains confined to a shift of the minimum. The minimum is still present when the wire is annealed and measured ill high vacuum. A heating of the wire by the measuring current cannot be the cause of the minimum. w 1_ Introduction. In two previous publications 1) 2) there was reported a minimum in the resistance-temperature curve of gold. This was not to be expected from the point of view of the electron theory of metals. The second publication 2) dealt with the influence of the chemical and physical impurities on the position of the minimum. The minimum shifted to higher temperatures for higher values of the residual resistance. All wires were annealed for some hours. The present paper concerns the influence of a very long annealing time ; one can expect that the physical impurities are smaller after a long annealing time, the chemical impurities are more homogeneously distributed over the pure metal and the gases are partially eliminated. Also the influence of the gases on the resistance curve and on the minimum is studied as it is not exactly known. In spectroscopic investigations of the purity of a metal the gases are imperceptible and so the quantities of gas in the metal are not known. w 2. Description o/the experimcnts. See w 2 of the former paper 2).
w 3. Results. The material used was obtained from W. C. H er a e u s (impurity < 10-4~ , traces Cu and Ag).
The wires were spiralized and suspended quite freely from the current-and potential wires in a glass mounting. All the wires were annealed in vacuum by a current going through them.