A method is presented for inferring atmospheric ozone information using infrared horizon radiance measurements in the 1042 cm−1 band. An analysis based on this method proves the feasibility of the horizon experiment for determining ozone information and shows that the ozone partial pressure can be determined in the altitude range from 50 down to 25 km. A comprehensive error study is conducted which considers effects of individual errors as well as the effect of all error sources acting simultaneously. The results show that in the absence of a temperature profile bias error, it should be possible to determine the ozone partial pressure to within an rms value of 15–20%. It may be possible to reduce this rms error to 5% by smoothing the solution profile. These results would be seriously degraded by an atmospheric temperature bias error of only 3K; thus, great care should be taken to minimize this source of error in an experiment. It is probable, in view of recent technological developments, that these errors will be much smaller in future flight experiments and the altitude range will widen to include from ∼60 km down to the tropopause region.