The mechanisms responsible for latitudinal and seasonal variations in the stratospheric NOy/N2O correlation, represented by the effective NOy yield from N2O loss, or FNOy, are explored using the Garcia-Solomon two-dimensional model. The model is run with and without Antarctic denitrification. Model results are compared to in situ NOy/N2O measurements taken onboard the NASA ER-2 high-altitude aircraft in the lower stratosphere during the 1994 Airborne Southern Hemisphere Ozone Experiment/Measurements for Assessing the Effects of Stratospheric Aircraft campaign, and to global-scale measurements taken onboard the Upper Atmosphere Research Satellite (UARS) from 1992 to 1993. The southern hemisphere midlatitude seasonal cycle observed by the ER-2 and the latitudinal gradients observed by UARS are consistent with the results of the denitrified model, although some aspects of the model results are sensitive to prescribed and/or calculated horizontal diffusion coefficients. The consistency with observations supports the model's prediction of a seasonal cycle in which FNOy increases at southern midlatitudes during winter due to descent of FNOy-enriched air from above and decreases in spring due to mixing with FNOy-depleted air from the denitrified polar vortex. Antarctic denitrification appears to affect midlatitudes mainly by a one-time dilution of the polar vortex following the final warming rather than by flow-through vortex processing during the winter. Because of the high concentrations of NOy at polar latitudes before denitrification a large fraction of total stratospheric NOy can be removed by a one-time dilution of the denitrified polar vortex. The nondenitrified model results generally do not agree well with observations, suggesting that denitrification strongly influences latitudinal and seasonal variations in FNOy in the southern hemisphere.¿ 1997 American Geophysical Union