A critical examination of the surface mobility mechanism of stress-cor
rosion cracking of Galvele is presented. Using arguments based upon th
e thermodynamics of stressed solids serious errors in the development
are elucidated and an accurate analysis of his ideas is presented. It
is shown that the original treatment of surface mobility overestimates
stress-corrosion crack velocities in ductile fee metals by about 14 o
rders of magnitude.