Abstract
The defect Raman bands in g-SiO2 may be related to the vibrations of bonds in a (100) cristobalite surface planar overlayer. A model which contains no dangling bonds is presented which is qualitatively complete and which is relevant to a wide range of experimental data.