Soil Respiration is typically characterized by measuring CO2 flux from the soil surface, but it can be determined by measuring the soil gas concentration and knowledge of the diffusive resistance to the soil surface. In either case, corrections should be made for the effect of temperature on gas solubility in soil water. Since CO2 is 30 times more soluble in water than O2, these corrections are 30 times larger for CO2 than for O2. Non-metabolic fluxes in CO2 also occur from changes in the bicarbonate equilibria. Because the gas/liquid partitioning of O2 is much less affected by temperature and not affected by bicarbonate equilibria, animal and human physiologists typically use measurements of O2 to quantify metabolic (e.g. VO2 max). Similarly, O2 fluxes provide a measurement of soil respiration that is more than 30 times less sensitive to temperature and pH changes than measurements of CO2. We examined the effect of temperature and barometric pressure on O2 and CO2 concentrations in a sterile sand medium using a Vaisala CO2 sensor and galvanic-cell O2 sensors that can resolve 6 ppm O2. We characterized the effect of temperature on the 4 factors that cause non-metabolic changes in soil gas concentrations: 1) partial pressure of O2, 2) water vapor dilution of the mole fraction of gas in soil, 3) gas/liquid partitioning, and 4) bicarbonate equilibria. When the appropriate correction factors are applied, O2 measurements provide an accurate, robust, and inexpensive measurement of soil respiration rates.