The electrical conductivity of pure water samples has been investigated with numerous experiments conducted in the years 1966 to 1971. Samples which had been submitted to an alternating voltage during 10 to 100 hours by means of a single electrode immersed in the water often showed after this treatment a conductivity which deviated significantly from that of untreated blanks. This conductivity change probably is not due to residual impurities with the exception of dissolved air. The conductivity deviation diminished progressively in number and amplitude and vanished in 1971. This disappearance suggests the influence of factors in the environmental conditions other than those controlled in a standard laboratory.