Several one-dimensional model experiments have been conducted in order to simulate the observed downward trend of total ozone and the shape of ozone profiles in the Antarctic during the austral spring. The results underline the fact that the O3 depletion is quadratically sensitive to the amount of active chlorine, which we have expressed as the fraction (F) of Clx that is in active form by polar sunrise. For F=1, this result is largely independent of the occurrence of heterogeneous chemistry after polar sunrise since all of the processing has already occurred. The main O3 loss is due to formation and photolysis of Cl2O2, the mechanism first suggested by Molina et al. <1987>. By comparison, using Brx levels similar to currently measured values, it was found that the BrO+ClO reaction plays a less important role in O3 depletion. The depth of the O3 minimum in spring appears to be related to the vertical extent of the region chemically processed by polar stratospheric clouds as well as the absolute levels of active Clx; more extensive vertical regions result in larger bite outs of the O3 profile. This may account for the difference in O3 minimum between 1986 and 1987. Simulations involving downward vertical transports are unable to reproduce observed decreases in total O3. We speculate that low trace gas mixing ratios of species such as N2O may be remnants of the previous summer and fall conditions, rather than signatures of downward transport during the springtime ozone depletion. ¿ American Geophysical Union 1990