The ozone long-term trends derived from the Stratospheric Aerosol and Gas Experiment II have been shown to be biased by the level of volcanic aerosol. We have investigated the spectral algorithm used to separate the partial optical thicknesses associated with ozone and aerosols. We present two possible causes of inaccuracies capable of producing relative errors on the slant path ozone optical thickness as large as 5--15%. The first one is related to the interpolation of the aerosol optical thickness at 0.6 μm which is not very well performed by solving the inverse problem of the unknown size distribution. The second error cause lies in the assumed altitude independence of the aerosol optical properties. Both error causes are discussed, and possible corrections are presented. ¿ 1998 American Geophysical Union