Moisture-induced delayed spallation (MIDS) of protective alumina scales at room temperature is a well known phenomenon. One mechanism proposes that water and Al-alloy react, produce hydrogen at the scale-metal interface, and enable spallation. To test this mechanism, preoxidized samples of a single crystal superalloy, Rene'N5 + Y, were subjected to standard cathodic hydrogen charging treatments known to produce hydrogen embrittlement in bulk Ni and Ni3Al alloys. Cathodic hydrogen charging, at < 1 mA and an estimated -0.45V SCE, stripped the scales at the oxide-metal interface, resulting in an initial loss of similar to 3 mg/cm(2) and little additional change with time. This was supported by macro-photos and SEM of the spalled surface. On the other hand. anodic polarization at < 1 mA produced less, but steady, linear weight loss (0.3 mg/cm(2)), primarily by anodic dissolution of the metal. Hydrogen charging was thus shown to be detrimental to the alumina scale-metal bond, supporting the hydrogen factor in MIDS. These and other MIDS results show remarkable similarities to embrittlement of Ni subject to hydrogen charging at similar potentials and varying amounts of interfacial (grain boundary) sulfur segregation. The MIDS phenomenon is also discussed in terms of comparative static corrosion fatigue characteristics. It is not necessarily related to other diverse moisture effects occurring at high temperature. Published by Elsevier Ltd.