Abstract The objective of stratigraphic velocity interpretation is to predict parameters such as lithology, fluid content, porosity, and formation pressure from seismic velocity data. Data from three sources are required to meet this objective: 1) stacked seismic sections; 2) seismic velocity profiles; and 3) well data. Interpreted seismic velocity profiles are combined with sequence boundary times marked on the stacked sections to compute sequence-keyed interval velocities. Interval velocities can be smoothed between sequence boundaries on a rational geological basis. Smoothed, sequence-keyed interval velocities are the basis for prediction of lithology and other parameters. These same velocities can also be used for migration, and time-depth conversion, as well as for determination of the low frequency component for inverse model studies. Well data, either from the immediate area or from a geologic look-alike elsewhere, are required to understand the geological significance of the seismic velocity data. Well data are used to determine interval velocity versus depth curves which reflect the normal compaction trends for various lithologies. These curves are applied to sequence-keyed seismic velocities to correct for normal compaction effects. Variations that remain after these corrections are mapped and used for prediction of lithology, porosity, fluid type, and formation pressure. Seismic velocity data, by themselves, often lead to ambiguous interpretation. However, comparison of velocity maps with seismic facies and structure maps and with other seismic parameters such as amplitude and continuity can lead to a final interpretation that is compatible with all available sources of information. An example of lithology prediction in the National Petroleum Reserve-Alaska is used to illustrate the major steps in the stratigraphic velocity interpretation procedure.