An assessment is made of the ability of current theory to explain the phenomenology of upper atmospheric ozone as revealed by the sizeable body of measurements presently available. The chemistry of ozone in the mesosphere and lower thermosphere is closely coupled to the chemistries of other oxygen/hydrogen-containing species, which must be considered concurrently. To provide insight into the sensitivity of model calculations to the choice of values used for key chemical rate constants and climatological parameters, simple analytical expressions for ozone are derived for those situations when it is in photochemical steady state, the mesosphere during daylight hours and the lower thermosphere throughout the full diurnal period. The model is found to reproduce the detailed Aladdin 74 rocket measurements of ozone from 50 to 95 km, numerous other measurements of mid-latitude ozone in the lower mesophere, the secondary maximum in the ozone vertical distribution at the mesopause, and the diurnal variability of ozone seen in the radio measurements of Wilson and Schwartz (1981) and other observations. The agreement with the Aladdin 74 data results from adjusting some key parameters within the uncertainties of laboratory measurements or known natural climatological variability. The variety of mid-latitude observations can be understood in terms of the estimated variability of environmental factors: diurnal, seasonal, and solar cycles in the solar illumination; the abundance of water vapor; and the details of the thermal profile of the atmosphere. The ozone secondary maximum results from the onset of the coupling between the active-hydrogen and active-oxygen chemistry and its observed variability may be a consequence of secular changes in mesopause dynamics. Above ~95 km, ozone observations are consistently higher than model results and cannot be accounted for by the set of reactions currently included in the model.