This thesis integrates data on deep sea gas hydrates from multichannel seismic (MSC) and other geophysical surveys with the results of deep sea drilling to estimate the distribution and amounts of deep sea gas hydrate and underlying free gas. The most important data are the velocity increase due to high-velocity gas hydrate and velocity decrease due to low-velocity gas, and core pore fluid chlorinity dilution due to hydrate dissociation. Data are available from the Deep Sea Drilling Project (DSDP) and Ocean Drilling Program (ODP), and numerous seismic surveys in three regions, on the Cascadia margin off Vancouver Island and Central Oregon, the Blake Outer Ridge off eastern USA, and the South American western margin off Chile. Hydrates are mainly detected through bottom-simulating reflectors (BSR) that mark the base of the hydrate stability field several hundred meters below the seafloor. The BSR is generally a simple symmetrical pulse of negative polarity that results from the impedance contrast between high-velocity sediments containing hydrate above the BSR and an underlying low-velocity zone containing free gas. There is usually no reflection from the top of the hydrate layer or from the base of the underlying gas so these boundaries are interpreted to be gradational. The hydrate and free gas concentrations have been estimated from surface seismic, ODP downhole sonic log, and vertical seismic profile (VSP) velocities, using models for how the velocity increases with hydrate concentration. DSDP and ODP core and log results have shown the sediment sections to be relatively homogeneous (on a seismic wavelength) in the areas studied, and the critical reference velocity-depth profile for no hydrate and no gas was obtained mainly from upward extrapolation of the deeper velocity data. High hydrate concentrations are inferred at all sites; between 15-50% of the pore space is filled with solid hydrates. For the Cascadia margin, MSC, sonic log, and VSP velocities are in excellent agreement in the 225 m thick interval above the BSR at ODP Site 889. At all sites, the hydrate concentration generally increases downward from about zero near the seafloor to a maximum just above the BSR. Evidence for low velocity free gas is seen in the velocity data of all three studied areas. At the Cascadia margin sites the low velocity zone underlying the BSR is thin (<30 m) and gas concentrations are estimated to be very low (<1%). Velocity results from the Blake Outer Ridge and the Chile margin indicate slightly higher gas concentrations, about 2.0%. Independent hydrate concentration estimates, based on simple hydrate dissociation interpretation of low chlorinity fluids in DSDP and ODP cores, give similar results in most of the studied areas. However, the reference pore fluid chlorinity before hydrate dissociation remains uncertain. P-wave and S-wave velocities, and Poisson's ratio for unconsolidated marine sediments have been computed using Biot (1956) effective fluid model and equations from Geertsma (1961). The results show that the effect of free gas on Poisson's ratio for the common approximately 50% porosity sediments is not as dramatic as was earlier predicted. The amplitude-versus-offset (AVO) response to high-velocity hydrate above the BSR and an underlying low-velocity gas layer in such sediments has also been modeled. In contrast to previous inference, no significant AVO effect is predicted even for quit high gas concentrations. The amount of methane trapped inside the hydrate structure is very large, about 164 m (super 3) of methane at STP for 1 m (super 3) of solid hydrate. If the estimated hydrate volumes of the Vancouver Island continental slope and the Blake Outer Ridge are representative of the worldwide distribution of hydrate, the worldwide estimated volume of methane in continental slope hydrate reservoirs is between 38-78X10 (super 15) m (super 3) (1.3-2.7X10 (super 6) TCF).