Transport inaccuracies in stratospheric models are a major source of uncertainty in predicting the environmental impact of trace gases with stratospheric source, such as emissions from high-flying aircraft. Because there are no observed tracers of stratospheric aircraft emissions, a direct evaluation of the models' transport of the emissions is not possible. Here we examine the relationship between the stratospheric residence time &tgr;R for tracers of midlatitude stratosphere source and the mean age &Ggr;, a transport diagnostic for which observations are available and have been used to evaluate stratospheric models. Employing a representation of the stratosphere that includes the basic kinematic features of the global circulation, we find that &tgr;R and &Ggr; are correlated over a range of plausible circulations, but that the correlation is imperfect. A major (but not sole) limitation on &Ggr; as a proxy for &tgr;R is the midlatitude tropopause height ZM. Elevating ZM reduces &tgr;R significantly, while &Ggr; is only weakly affected. The relationship between &tgr;R and &Ggr; seen among the participants of a recent stratospheric model intercomparison is consistent with this analysis. ¿ 2000 American Geophysical Union