Local ozone loss rates due to the ClO+ClO and BrO+ClO cycles are calculated using ClO, pressure, and temperature from in situ aircraft measurements and representative BrO mixing ratios. Ozone loss during the vertical profiles executed by the ER-2 near 72 ¿S usually extended over a deep altitude range rather than reaching a maximum at the top of the profiles. This is due to the strong pressure dependence of the rate determining steps. In the Antarctic, very high ozone loss rates (>5 ⋅ 106 cm-3 s-1) were observed at altitudes with potential temperatures below 400 K, where advective exchange is likely to be much more rapid than at higher altitudes. On September 22, 1987, the ER-2 measured an ozone loss rate of about 2.8 Dobson units (DU) per 12 sunlit hours in the 350--400 K range and 2.0 DU in the 400--450 K range near 72 ¿S. Rapid ozone loss in the Arctic did not extend below 400 K in the available data. The calculated average loss rate, which is nonlinear, in general depends on the order in which the terms are averaged. Loss rates calculated by averaging the ClO, pressure and temperature for up to 2400 s (about 500 km) generally agree with the average of the local loss rate to within one percent except at the edge of the vortex, where the difference can be up to 30%. Adiabatic temperature changes of up to 10 K change the ozone loss rate by less than 10% because the temperature and pressure effects nearly cancel. Thermal decomposition of Cl2O2 was not important along sunlit portions of ER-2 flight tracks if equilibrium is assumed between ClO and Cl2O2. The effect of recalibration of the ClO data on the calculated loss rates is discussed.