Research in 1980/81 was carried out in three areas: (i) electron diffraction experiments on diffusion bonded (welded) boundaries; (ii) combined EBIC and high voltage transmission electron microscopy in order to investigate the correlation between structure and electrical properties and (iii) passivation of the electrical activity of crystal defects with hydrogen. The diffraction experiments did not furnish new information on the structure of grain boundaries in silicon, but were valuable as a check on the previous findings. In addition, they offered the possibility to separate double diffraction effects from boundary scattering. The combined EBIC and HVTEM experiments established (for the first time) unambiguously that coherent twin boundaries per se are not electrically active. The localized electrical activity observed in coherent twin boundaries is due to the presence of intrinsic dislocations, which are partial dislocations of the Schottky type. However, not all partial dislocations studied were electrically active, for reasons which are not completely understood. The combined EBIC and HVTEM investigations showed further (again for the first time) that many of the linear boundaries which were previously assumed to be coherent twin boundaries are second order twins, and that these twins are strongly electrically active. In passivation experiments, it is necessary to separate influence of the heat treatment from those of hydrogen. Passivation reduces the activity of some, but not of all defects. Passivation is particularly effective in reducing the electrical activity of deformation induced dislocations, which conceivably have a different core structure from grown-in dislocations.