The major features of the individual velocity models, Poisson’s ratio values, and crustal complexity derived from the interpretation of seismic data sets from four long-range seismic refraction – wide-angle reflection experiments are summarized. The experiments were conducted from 1982–92 in the southeastern portion of the Canadian Shield. In the conventional analysis of seismic refraction – wide-angle reflection data, only the onset times and amplitudes of the major arrival phases are used to derive seismic velocity models of the region under study. These models are over smoothed, have a number of intermediate discontinuities, are unable to explain the Pg coda, and bear very little resemblance to the models derived from the analysis of near-vertical seismic reflection data. In this paper some of the differences between seismic models derived from near-vertical reflection analysis and those from refraction analysis are reconciled from an analysis of the wide-angle reflection fields of the crustal coda waves that follow the first arrivals. This was done using a migration technique that to a first approximation maps the amplitudes of the record sections into a two-dimensional (2-D) complexity section. These new sections show significant lateral variations in crustal and Moho reflectivity and may be used to complement the 2-D velocity anomaly sections and near-vertical reflection sections. The method was based on a numerical study that showed that the coda can be explained with a class of complex heterogeneous models in which sets of small-scale, high-contrast sloping seismic reflectors are “embedded” in a uniform seismic velocity gradient field.