GENERAL CIRCULATION OF THE LOWER STRATOSPHERE 393 Refs. 1 and 2). In the troposphere temperatures decrease with altitude and latitude in both seasons, and zonal winds have maximum westerly components in middle latitudes at heights of about 12 km. The winter valuesare the largest. Above the maximums, zonal winds decrease to a transition zone at about 25 km. Temperatures in the 15- to 25-km region vary little with height in middle latitudes but increase polewards to about 60° latitude in winter and90° in summer. Above 25 km, temperatures increase with altitude to about 50 km and decrease again to 80 km. In winter there is a decrease, and in summer there is an increase polewards in the 25- to 50-km region. The reverse occurs in the 50- to 80-km region. Associated with this temperature struc- ture is another global scale vortex that is westerly in the winter hemisphere and easterly in the summer hemisphere. In addition to these vortexes, there is yet another in the equatorial stratosphere which alternates between westerly and easterly with a 26-month period. One of the major problems of general-circulation research at the present time is to explain why these particular configurations of temperature and wind occur. The problem has been attacked chiefly by diagnostic studies of the momentum and the energy budgets. Also of particular value have been the concepts of available potential energy as formulated by Lorenz.‘ If we consider the diabatic heating and cooling of the atmosphere by radiative processes, we have the following general picture (based on work by Davis® for the upper troposphere, 5 to 15 km; by Kennedy® for the 10- to 30-km region; and by Murgatroyd and Goody’ for the 30- to 80-km region): The low-latitude troposphere is heated by radiation while the higher latitudes are cooled; thus there is a generation of available potential energy. This in turn is converted into kinetic energy by large-scale quasi-horizontal eddy processes that transport heat polewards. Incidental effects are that angular momentum is also systematically accumulated in the upper troposphere in middle latitudes and a westerly jetstream is built up. The energy-releasing processes are essentially adiabatic motions in which the mean slope of air parcels are about one-half as steep as the mean inclination of potential-temperature surfaces, as pointed out by Eady.® In the lower stratosphere, as noted elsewhere,’"! the situation is different, with the mean slopes of individual parcels being steeper than the mean isentropic surfaces. Therefore a forced motion, with kinetic energy being converted to potential energy, is implied. As noted at the previous AEC fallout conference,’? the possibility that poleward-moving parcels are descending and warming adiabatically while equatorward-moving parcels are rising and cooling adiabatically allows one to explain the observations of White,!? Peixoto,'4 Murakami,!® and Peng'® of a countergradient heat flux in the lower stratosphere. It is perhaps worth emphasizing that the concept of these sloping motions in the stratosphereinitially was developed from rather