A wide–angle reflection/refraction survey, conducted near Yellowknife, N.W.T., Canada has revealed a complex crust in contrast to the simple model indicated by a large–scale refraction survey. A near–vertical reflection survey, conducted in north–central British Columbia, using dynamite detonated in a shallow lake, recorded coherent energy at two–way times of up to 11 sec. A reflector at a depth of 30–35 km is thus indicated which corresponds to a refraction–determined Moho depth for the area. CRP stacking attempts have been unsuccessful for both of these surveys due mainly to the complex overburden encountered but possibly due in part to the nature of the deep reflectors themselves. A reflector comprised of a series of thin layers of alternating high and low velocities will produce a composite reflected pulse, whose amplitude, frequency content, and apparent arrival time will change as the reflecting angle changes. Successful application of the CRP method would require restricting its use to a much narrower range of incident angles than we have employed. A Vibroseis survey, duplicating the coverage of the B.C. reflection profile, indicated the same reflector at 11 sec. The Vibroseis section remains superior despite exhaustive attempts at enhancement of the dynamite–produced data.