We describe statistical comparisons between Polar Ozone and Aerosol Measurement (POAM) III and Stratosphere Aerosol and Gas Experiment (SAGE) II measurements of aerosol extinction in 1998 and 1999. SAGE II and POAM III are in qualitative agreement, and show that since the launch of POAM III in March of 1998, stratospheric aerosol extinctions at visible and near-IR wavelengths have remained at background levels. We present quantitative differences between the SAGE II and POAM III extinctions at 1.02 μm and 0.45 μm for temporally and spatially coincident measurements. At 1.02 μm the instruments agree to within about ¿30% from 10 to 22 km, where most of the aerosol extinction lies. Differences at 0.45 μm are similar to 1.02 μm in the southern hemisphere, but much larger in the northern hemisphere. We show that differences between SAGE II and POAM III slant path optical depths closely resemble the aerosol extinction differences at 1.02 μm, but not at 0.45 μm. On the basis of these results we conclude that the aerosol extinction differences at 1.02 μm arise from differences in the slant path optical depth measurements of the two instruments, not from species separation differences in the retrieval algorithms. However, we tentatively attribute the aerosol extinction differences at 0.45 μm to an error in the NO2/aerosol species separation in the SAGE II retrievals. We present theoretical simulations incorporating the maximum plausible systematic pointing or timing errors in the POAM III instrument. We conclude that these errors cannot explain the 1.02 μm optical depth or aerosol extinction differences, although it is possible that they contribute to the observed differences. We thus attribute these differences to fundamental radiometric differences between POAM III and SAGE II that remain to be resolved. We also compare derived surface areas and volume densities from POAM III, SAGE II, and the Halogen Occultation Experiment (HALOE). The statistical differences can be explained by the aerosol extinction differences in the SAGE II comparisons, and by the lack of sensitivity on the part of POAM III and SAGE II to very small aerosols in the HALOE comparisons. Nevertheless, we show qualitative agreement between all three instruments. The results presented here show that the POAM III aerosol measurements are valid for scientific studies. ¿ 2001 American Geophysical Union