We have investigated the impact of high resolution, temperature-dependent CO2 cross-section measurements (Lewis and Carver, 1983) on calculations of photodissociation rate coefficients in the Martial atmosphere. Since the cross-section measurements exhibit large variations on a 10--20 ¿ scale, the calculations are sensitive to the wavelength resolution. We find that the adoption of 50 ¿ intervals for the purpose of computational efficiency results in errors in the calculated values for photodissociation of CO2, H2O, and O2 which are generally ≤10%, but as large as 20% in some instances. These are acceptably small errors, especially considering the uncertainties introduced by the large temperature dependence of the CO2 cross-section. The inclusion of temperature-dependent CO2 cross-sections is shown to lead to a decrease in the diurnally averaged rate of CO2 photodissociation as large as 33% at some altitudes, and increases of as much as 950% and 80% in the photodissociation rate coefficients of H2O and O2, respectively. The actual magnitude of the changes depends on the assumptions used to model the CO2 absorption spectrum at temperatures lower than the available measurements, and at wavelengths longward of 1970 ¿. Given our current understanding, these results minimize or even reverse the sense of the CO2 chemical stability problem. ¿ American Geophysical Union 1993