In December, 1934, Mr. R. Sussex extruded some silver lead alloy wires in order to determine their electrical resistance. After the determinations had been made each wire was threaded through a sheet of paper on which the composition was noted. The wires were placed in a more or less tangled mass in a cupboard in March, 1935. It was not until September, 1936, that they were re-examined, and it was found that during the eighteen months some of the wires had corroded very badly. It was not until March, 1938, that the wires and the corrosion product could be examined, and the results are given in the following pages. It is of interest to note that none of the numerous silver-lead alloys which have been used for creep testing in the rolled condition have shown a similarphenomenon. The silver content of the rolled alloys was, however, never greater than 01%.GENERAL APPEARANCE OF THE CORRODED WIRESThe wires contained 01% up to 28% of silver, and a general examination showed that the corrosion had progressed farther the higher the silver content. This can be seen in Fig. 1, which is a photograph taken in March, 1938, of typical wires at a 'magnification of five diameters. In the early stages the corrosive action seems to be localized and an area of grey, yellow or reddish brown powder appears. The grey corrosion product is by far the one, most frequently seen, and in the later stages the only one visible from the outside. The reddish coloured product is sometimes seen on the residual metallic centre when the grey product has been broken away.The irregular extent of the corrosion can be seen in Fig. 2, which shows (natural size) the 5% silver wire after standing two years. The corroded parts are seen to be much thicker than the slightly corroded ones. The corrosion product, being non-metallic, is quite brittle and falls away from the wire on bending. In the extreme case the' whole wire has corroded and presents a cracked appearance as seen in Fig. 1 (281% silver).An approximate idea of the extent of corrosion can be obtained by measuring the diameter of the corroded part. The original diameter in each case was 050/ 051 inches and in Table I are given the diameters of the corroded parts after Hand 3 years respectively. When the diameter...