More systematic observations of airflow in the lee of mountain ridges are required for the appraisal of theoretical work, for forecasters to gain experience of the relation of the flow to the synoptic situation for the inclusion of information in flight forecasts, for the planning of flights in mountainous regions, and for the planning of research in chosen areas. To assist in the recording of phenomena, waves are classified into three types: (i)Downwind wave: a series of lee waves whose intensity decreases with altitude, quasi-stationary regions of strong turbulence depending on the mountain height, often with a periodic regeneration described below; produced by the lee slope. (i)Obstacle wave: a smooth deformation above the obstacle, common when there is a pronounced inversion with wind increasing with height above; no turbulence; often accompanied by continual release of thermals from the windward slope which produces it. (i)Composite wave: combination of (i) and (ii). Conditional instability at higher levels seems important for this; more frequent in regions of high mountains. Clouds are good indicators of wave crests but, if air is saturated throughout, or if there are no clouds, observations must be made using a variometer. In a downwind wave, the low-level rotor-like clouds and the upper bands of billow clouds often change their position over fairly short periods. They move slowly downwind and then make a quick leap against the wind, on account of the periodic release of eddies with a nearly stable flow-pattern in between. At the same time periodic changes of surface wind and pressure are observed at places in the downwind region. These changes can be thought of as the periodic release of circulations from the lee slope, each eddy passing through alternate stages of motion and rotation. The motion becomes insignificant at about one wavelength from the ridge, where the rotation may be seen in rotor clouds. The rotation then suddenly weakens and the eddy increases its horizontal motion, the rotation becoming dominant again, for a short time, in the position of the second wave, and this may be repeated several times. Sometimes successive eddies maintain a quasi-stationary pattern of flow in the lee of the ridge. The temperature is less in the wave crest and greater in the wave trough than in between, where it is equal to that of the undisturbed stream. These temperature anomalies can be measured by glider if the system is stationary, if the flights are planned to make measurements in ascending and horizontally moving air during the ascent, and in the downcurrent during a rapid descent from the greatest height reached. They can also be measured by radiosondes released either simultaneously or in quick succession from suitably chosen sites downwind of the ridge. The amplitude of the wave can be deduced from the temperatures measured. The use of specially designed blank forms for the reporting of wave phenomena would help. This is particularly true of soaring flights which have been and will be the main source of information provided that pilots have suitable means for recording the phenomena. Since soaring flights exploit the conditions in question it is better to obtain as much information as possible from them as they occur, rather than by short-period investigations.