The Wind and Temperature Spectrometer (WATS) measured the in situ neutral winds, the neutral particle temperatures, and the concentrations of selected gases. The objective of this investigation was to study the interrelationships among the winds, temperatures, plasma drift, electric fields, and other properties of the thermosphere that were measured by this and other instruments on the spacecraft. Knowledge of how these properties are interrelated contributed to an understanding of the consequences of the acceleration of neutral particles by the ions in the ionosphere, the acceleration of ions by neutrals creating electric fields, and the related energy transfer between the ionosphere and the magnetosphere. Three components of the wind, one normal to the satellite velocity vector in the horizontal plane, one vertical, and one in the satellite direction were measured. A retarding potential quadrupole mass spectrometer, coupled to the atmosphere through a precisely orificed antechamber, was used. It was operated in either of two modes: one employed the retarding capability and the other used the ion source as a conventional nonretarding source. Two scanning baffles were used in front of the mass spectrometer: one moved vertically and the other moved horizontally. The magnitudes of the horizontal and vertical components of the wind normal to the spacecraft velocity vector were computed from measurements of the angular relationship between the neutral particle stream and the sensor. The component of the total stream velocity in the satellite direction was measured directly by the spectrometer system through determination of the required retarding potential. At altitudes too high for neutral species measurements, the planned operation required the instrument to measure the thermal ion species only. A series of four sequentially occurring "slots" --each a 2-s long measurement interval-- was adapted for the basic measurement format of the instrument. Different functions were commanded into these "slots" in any combination, one per measurement interval. Thus the time resolution can be 2, 4, 6, or 8 seconds. Further details are found in N. W. Spencer et al., Space Sci. Instrum., v. 5, n. 4, p. 417, 1981.