An experimental investigation has been made of the factors which affect the velocity of vibratory signals in porous media. It is shown from the results of experiments carried out on appropriate natural and synthetic porous systems that the time-average formula previously suggested by Wyllie, Gregory, and L. W. Gardner is of considerable utility. The effects are examined of differential compacting pressures on the applicability of this formula to consolidated and unconsolidated rocks. It is shown that the time-average relationship cannot be applied to determine the total volumetric porosity of carbonate rocks which are vugular and fractured. In such rocks, paradoxically, this circumstance may be advantageous because of the lithological information that may be obtained from an appropriate combination of velocity and nuclear log data. The effects of oil and gas saturation on velocity have been examined experimentally and are found to be comparatively minor. The combination of velocity data with information from electric logs in order to locate zones of oil and gas saturation is shown to be generally valuable; this is particularly so when holes are drilled with oil-base mud. Some discussion is given of the possible effects on velocity measurements of the relative wettability of rock surfaces by various liquids. Owing to instrumental limitations, it cannot necessarily be assumed that measurements made in the laboratory are directly applicable to the interpretation of velocity data obtained under field conditions.