During Ocean Drilling Program Leg 204, 80 in situ measurements of subseafloor temperature were made; 68 of these showed good frictional pulses on insertion and extraction from the seafloor and were used to constrain the subsurface temperature. Considering uncertainties from various sources, uncertainties in the in situ temperatures are estimated to generally be less than + or -0.3 degrees C. The data are consistent with a purely conductive temperature regime at all sites, and there is no resolvable difference in heat flow between sites on the flanks of southern Hydrate Ridge and sites near the summit, where other data indicate that free gas is venting into the ocean, gas hydrate is forming rapidly, and free gas and gas hydrate coexist in the sediments. We interpret this apparent paradox to indicate that the aqueous fluid flow is decoupled from free gas flow and that advection of free gas does not have a significant effect on the temperature field. The temperature data, which include several measurements within a few meters of the predicted base of the methane hydrate stability field (calculated for the measured pore water salinity at each site) also indicate that the bottom-simulating reflection (BSR) corresponds to the base of gas hydrate stability within measurement uncertainties, although a systematic shallowing of the BSR by as much as 10 m is possible. The heat flow indicated by the Leg 204 measurements and the regional BSR depth is significantly lower than the heat flow predicted based on the age of the subducting plate and the thickness of the accretionary complex. Several measurements made near the summit at depths shallower than 60 meters below seafloor show anomalous behavior consistent with low in situ thermal conductivity, possibly because of the presence of free gas and/or massive gas hydrate in these sediments.