Representative reduction of crystallographic orientation data

Two conceptually different methods to representatively reduce large experimentally measured orientation data sets are derived and discussed. Examples are given which show that, with approximately 3% of the experimental data, the crystallite orientation distribution can be very well reproduced.