Stratospheric ozone profiles measured by the Global Ozone Monitoring Experiment (GOME on ERS-2) and Stratospheric Aerosol and Gas Experiment II (SAGE II on ERBS, data version 6.2) were compared over a 4-year time period (1996--2000). GOME measures the reflected and backscattered radiation from Earth, and vertical profiles are derived from nadir observations using the Full Retrieval Method (FURM, version 5.0), which is based upon an advanced optimal estimation inversion scheme; SAGE II uses solar occultation to measure vertical profiles of ozone with an instantaneous vertical field of view of 0.5 km at the Earth limb. Coincident measurements are identified by limiting time differences and distance between two observation points. Since for ozone in the lower stratosphere gradients in the horizontal distribution evolve from transport processes, validation at the border of different air masses is rendered more difficult. Important factors influencing the ozone distribution are the tropopause height and whether a measurement is taken within or outside the polar vortex. This was taken into account when the validation was reduced to matches where both measurements were within the same air mass. Overall, comparisons show that between 19- and ~34-km good agreement between the ozone profiles of the two satellite instruments is achieved. If lower stratospheric ozone is strongly depleted during polar spring, a homogeneity condition has to be imposed on the GOME and SAGE II measurements by requiring an upper limit on the potential vorticity difference at the 475-K isentrope.