This velocity study is limited to data from one well in South Texas. Two short-interval velocity logging methods compared with conventional seismic geophone data show large discrepancies. The Shell short-interval velocity log agrees within close limits to the conventional seismic data except for the lower 4,000 ft. The indicated delay times for the upper 2,000 ft of this 4,000-ft interval are short by 6.5 percent, and indicated delay times for the lower 2,000 ft are short by 4.0 percent. The Schlumberger Sonic Velocity Log, limited in this survey to the bottom 4,200 ft of hole, indicated delay times larger than the seismic time by more than 5 percent. There is a difference of approximately 9 percent between the two velocity logs, even though the tools were of similar dimensions. The spacing between detectors was three feet, and the distance from transmitter to near receiver was four feet for the Shell tool and three feet for the Schlumberger tool. An analysis of the basic data is necessary to resolve these discrepancies. There is no check on the Sonic data in its present form, but a thorough study of the Shell Oscillogram log and conventional seismic data for errors fails to explain the 6.5-percent and 4-percent discrepancies in the Shell short-interval velocity data. The conclusion must be drawn that these discrepancies are real. This survey demonstrates the necessity to check short-interval velocity logging with conventional seismic shots to maintain acceptable seismic well velocity standards.