[en] Complete text of publication follows. The intense absorption spectrum of the hydrated electron, with maximum at 720 nm in room temperature water, shifts strongly to the red as the temperature is raised. In order to use this strong signal for dosimetry and calculation of second order reaction rates, the extinction coefficient must be measured as a function of temperature (and also pressure in supercritical water). Determination of the absolute extinction coefficient of a transient is a difficult undertaking. We have used two pulse radiolysis methods. In the first, we directly observe the kinetics of hydrated electron scavenging by methyl viologen, to form the strongly colored MV⁺ cation. The extinction coefficient of this long-lived radical was carefully measured by electrochemical and temperature cycling many years ago, so the measured ratio of hydrated electron absorption to MV⁺ absorption gives the desired extinction coefficient of (e-)_aq. The MV⁺ spectrum has only been quantitatively measured up to 200 deg C, so the method is not valid at higher temperature. In the second method, transient absorption of hydrated electron is recorded as it is being scavenged by N₂O or SF₆. Direct comparison of the transient absorption with the measured (N₂ or F^-) product yield allows the extinction coefficient to be calculated. While the purpose of our study was to investigate high temperature water, we were astonished to discover that the room temperature hydrated electron extinction coefficient has been incorrectly reported (low) by 10%. We can now demonstrate how this error arose in calibration of pulse radiolysis yields. It has been remarked for decades that the integrated oscillator strength of the hydrated electron is less than unity. Our new measurements rectify this problem for room temperature, and we will present results for high temperature water.