Accurate knowledge of engine exhaust gas temperatures is important for engine design and diagnostic efforts. To survive the harsh exhaust gas environment, most practical devices for determining exhaust gas temperatures are relatively large. Because of their size, these devices do not provide instantaneous gas temperatures but provide a single equilibrium temperature. This investigation compared computed thermocouple equilibrium temperatures of the exhaust gas to computed time-averaged and mass-averaged gas temperatures for nine different engine conditions. Mass-averaged gas temperatures are directly related to the exhaust gas energy whereas thermocouple and time-averaged gas temperatures are not directly related. For most of the engine conditions of this study, the exhaust gas energy based on time-averaged gas temperatures was about 10% lower than the energy based on mass-averaged temperatures and the energy based on thermocouple temperatures depended on the thermocouple properties.The mass-averaged, time-averaged and thermocouple exhaust gas temperatures varied as a function of engine conditions. For a spherical thermocouple with a diameter of 0.076 cm and an emissivity of 0.85, the computed thermocouple equilibrium temperature was about 20 K lower than the time-averaged exhaust gas temperature for variations in engine speed, spark timing and equivalence ratio. Variations in engine load did not appreciably change the thermocouple equilibrium temperature, however the time-averaged gas temperature decreased for increasing engine loads.The thermocouple temperature for a fixed engine condition was examined as a function of thermocouple emissivity, diameter and shape. The calculations resulted in thermocouple temperatures which increased for decreasing values of emissivity for all thermocouple diameters and for decreasing values of the thermocouple diameters for moderate to high emissivities. For low emissivities, decreasing thermocouple temperatures were obtained for decreasing values of the thermocouple diameter.