It has long been recognised that tbe application of dip-moveout (DMO) does not necessarilyresolve all stacking velocity conflicts in areas where: - there is a velocity gradient - where there are steep dips Such circumstances may require pre-stack time migration. A problem with full pre-stack time migration is knowledge of the velocities with which to drive the migration. In principle finding out could be a very expensive exercise. The following case-history demonstrates how to obtain high quality from an efficient approach to gaining a good image from pre-stack time migrated data at an optimum velocity.