A direct-sampling, mass-spectrometric technique has been used to measure simultaneously the solubilities of He, Ne, Ar, Kr, and Xe in fresh water and NaCl brine (0 to 5.2 molar) from 0° to 65 °C, and at 1 atm total pressure of moist air. The argon solubility in the most concentrated brines is 4 to 7 times less than in fresh water at 65 °C and 0°C, respectively. The salt effect is parameterized using the Setschenow equation. ln [ β i o(T) β i(T) = MK i M(T) where M is NaCl moiarity, β i o ( T) and β i ( T) the Bunsen solubility coefficients for gas i in fresh water and brine, and K i M ( T) the empirical salting coefficient. Values of K i M ( T) are calculated using volumetric concentration units for noble gas and NaCl content and are independent of NaCl molarity. Below about 40°C, temperature coefficients of all K i M are negative. The value of K He M is a minimum at 40°C. K Ar M decreases from about 0.40 at 0°C to 0.28 at 65 °C. The absolute magnitudes of the differences in salting coefficients (relative to K Ar M ) decrease from 0° to 65°C. Over the range of conditions studied, all noble gases are salted out, and K He M ≲ K Ne M < K Ar M < K Kr M < K Xe M . From the solubility data, we calculated Δ G 0 tr , Δ S 0 tr , Δ H 0 tr and Δ C O p,tr for the transfer of noble gases from fresh water to 1 molar NaCl solutions. At low temperatures Δ S 0 tr , is positive, but decreases and becomes negative at temperatures ranging from about 25°C for He to 45°C for Xe. At low temperatures, the dissolved electrolyte apparently interferes with the formation of a cage of solvent molecules about the noble gas atom. At higher temperatures, the local environment of the gas atom in the brine appears to be slightly more ordered than in pure water, possibly reflecting the longer effective range of the ionic fields at higher temperature. The measured solubilities can be used to model noble gas partitioning in two-phase geothermal systems at low temperatures. The data can also be used to estimate the temperature and concentration dependence of the salt effect for other alkali halides. Extrapolation of the measured data is not possible due to the incompletely-characterized minima in the temperature dependence of the salting coefficients. The regularities in the data observed at low temperatures suggest relatively few high-temperature data will be required to model the behavior of noble gases in high-temperature geothermal brines.