Abstract: Both O₃ and CO₂ in the stratosphere have unusual isotopic signatures in that they are more enriched in 17O than is expected based on their enrichments in 18O and the usual 'mass-dependent' relationship between 17O and 18O. This 17O isotope anomaly in O₃ is due to non-mass-dependent isotope effects in O₃ formation, while the 17O isotope anomaly in CO₂ is thought to arise, at least in part, from ultraviolet photolysis of anomalously enriched O₃, followed by isotope exchange between the resulting O(1D) and CO₂. However, efforts to quantify the transfer of the isotopic signature from O₃ to CO₂ have so far been unsuccessful. Furthermore, previous laboratory experiments have been interpreted to indicate that there may be additional non-mass-dependent isotope effects both in O ₃ photolysis and in the CO₂+O(1D) isotope exchange reaction. To address these issues, a photochemical kinetics model was developed to simulate all isotope-specific reactions involved in the isotope transfer. Using this model to simulate earlier O₃ photolysis experiments revealed that non-mass-dependent isotope effects in O₃ photolysis are not required to reproduce the experimental data, nor are recent theoretical calculations inconsistent with the laboratory results as they previously appeared to be. The model was then modified to include CO₂ and used to test the sensitivity of the CO₂ isotopic composition to isotope effects in underlying reactions. The predictions of this model were tested by performing a series of laboratory photochemistry experiments in which the isotopic composition of CO₂ in a mixture with O₂ was directly measured after irradiation with UV light. The results show that the reaction scheme in the model successfully describes the chemical origin of the 17O isotope anomaly in CO₂, implying that no non-mass-dependent isotope effects exist in CO₂+O(1D), O₃ photolysis, or any other reaction besides O₃ formation. These findings may lead to a deeper understanding of the chemical physics of non-mass-dependent isotope effects in general. In addition, measurements of the anomaly in CO ₂ in the atmosphere and in O₂ in ice cores can now be more confidently used as tracers of stratospheric ozone chemistry and transport and of biospheric productivity on annual and millennial timescales.