A comprehensive model for the Martian upper atmosphere, consistent with results obtained in the recent Mariner IV occultation experiment, is developed. It is argued that the ionosphere detected in this experiment should not be interpreted as an F2 layer analogous to the ionosphere. The possibility of an Fl interpretation is explored, and attention is directed to important difficulties associated with this interpretation It cannot, however, be categorically excluded at this time. A detailed E region model is derived and shown to compare favorably with observation. The accuracy of thermal calculations is discussed. In particular it is argued that eddy transport does not provide an important mode of heat loss from the Martian thermosphere. It is concluded that 4000 K represents a reasonable lower bound to the exospheric temperature on Mars. The photochemistry of CO2 is discussed briefly and attention is directed to the possible importance of reactions involving O(1D). It is suggested that association of O(1D) and CO may occur rapidly, and as a consequence the upper atmosphere may be deficient in dissociation products such as CO, 02, and 0.