Abstract
It is argued that the viscosity of vitreous silicon dioxide must be related to the motion of defects. These are thought to be charged and to comprise a threefold-coordinated silicon atom together with a non-bridging oxygen atom resulting from a broken bond. A single broken bond can lead to plastic flow through its capacity to change the number of bonds in a -Si—O–Si—O– ring. An alternative mechanism involving two broken bonds is outlined. Following the experimental results of Hetherington et al. (1964), it is argued that both mechanisms may be operative in appropriate ranges of temperature.