Visible spectra of the zenith-sky were recorded at Faraday in Antartica during 1990. Line-of-sight NO2 amounts were deduced by fitting laboratory cross-sections to the spectra. A new procedure was used to determine the offset to the zero of the measurement from the measurement series itself. This offset includes the effect of the reference spectrum—a single reference was used for the whole data set—plus any instrumental or analytic artefacts. It is important to determine this offset to the zero because it affects the absolute calibration of the zero of the vertical amount of NO2. In the new procedure, offsets were determined from fits to plots of the measured line-of-sight NO2 vs calculated air-mass factors, whose curvature had been precomputed from model predictions rather than the straight-line fits of Langley plots. The resulting vertical NO2 amounts were significantly lower than those deduced using a more subjective determination of offset, particularly in the early spring when denitrification might be expected.