The paper discusses the use of Eliassen-Palm (EP) cross sections as a diagnostic tool for analyzing large-scale, quasi-geostrophic motion in the troposphere. EP cross sections show the Eliassen-Palm flux \( F \) by arrows and its divergence by contours. The divergence of \( F \) reflects the magnitude of transient and irreversible eddy processes at each height and latitude, proportional to the northward flux of quasi-geostrophic potential vorticity. The direction of \( F \) indicates the relative importance of heat and momentum eddy fluxes. If the eddy dynamics is Rossby wavelike, \( F \) also measures net wave propagation. Observational and theoretical EP cross sections for the layer 1000-50 mb are presented and discussed. The observational cross sections for transient eddies are more reliable than those for stationary eddies and resemble nonlinear planetary instability simulations. They do not match linear instability theory or realistic mean state specifications. The EP cross sections for stationary eddies in winter show a significant dynamical linkage, while those in summer are different but less well determined. The "residual meridional circulations" associated with the observed EP cross sections are also discussed. The paper reviews the theoretical background, including the Eliassen-Palm theorem and its nonlinear extension, and the relationship between \( F \) and the eddy-induced forcing of the mean state. It also discusses the connection between \( F \) and wave theory, showing that \( F \) is parallel to the meridional projection of the local group velocity under certain conditions. The data sources and graphical conventions for calculating EP cross sections are described, and the results are compared with theoretical models of baroclinic waves. The EP cross sections for transient eddies and the life cycle of a nonlinear baroclinic wave are presented, highlighting the interplay of linear and nonlinear processes. The time-averaged picture of the nonlinear wave life cycle is closer to the observed EP cross sections than any of the linear models. The paper concludes by suggesting that EP cross sections are useful for understanding the dynamics of the troposphere and for assessing eddy parameterizations.The paper discusses the use of Eliassen-Palm (EP) cross sections as a diagnostic tool for analyzing large-scale, quasi-geostrophic motion in the troposphere. EP cross sections show the Eliassen-Palm flux \( F \) by arrows and its divergence by contours. The divergence of \( F \) reflects the magnitude of transient and irreversible eddy processes at each height and latitude, proportional to the northward flux of quasi-geostrophic potential vorticity. The direction of \( F \) indicates the relative importance of heat and momentum eddy fluxes. If the eddy dynamics is Rossby wavelike, \( F \) also measures net wave propagation. Observational and theoretical EP cross sections for the layer 1000-50 mb are presented and discussed. The observational cross sections for transient eddies are more reliable than those for stationary eddies and resemble nonlinear planetary instability simulations. They do not match linear instability theory or realistic mean state specifications. The EP cross sections for stationary eddies in winter show a significant dynamical linkage, while those in summer are different but less well determined. The "residual meridional circulations" associated with the observed EP cross sections are also discussed. The paper reviews the theoretical background, including the Eliassen-Palm theorem and its nonlinear extension, and the relationship between \( F \) and the eddy-induced forcing of the mean state. It also discusses the connection between \( F \) and wave theory, showing that \( F \) is parallel to the meridional projection of the local group velocity under certain conditions. The data sources and graphical conventions for calculating EP cross sections are described, and the results are compared with theoretical models of baroclinic waves. The EP cross sections for transient eddies and the life cycle of a nonlinear baroclinic wave are presented, highlighting the interplay of linear and nonlinear processes. The time-averaged picture of the nonlinear wave life cycle is closer to the observed EP cross sections than any of the linear models. The paper concludes by suggesting that EP cross sections are useful for understanding the dynamics of the troposphere and for assessing eddy parameterizations.