Entry of heated solar wind plasma into the magnetosphere is examined using electron distributions measured by AMPTE UKS and HEOS 2. In particular, the angular structure of the electron distributions is studied within the transition region separating the magnetosheath from the inner magnetosphere. The measurements suggest that electrons in the outer part of the transition region originate in the magnetosheath, whilst the population closer to the Earth consists of electrons from the magnetosphere combined with an energized magnetosheath component. This energized component contains “counterstreaming” electrons, which are confined to directions closely parallel and antiparallel to the magnetic field direction. The possibilities, that the energization of the counterstreaming electrons is cumulatively gained from either waves, electric fields perpendicular to the magnetic field, or quasi‐Fermi acceleration, are discussed. It is not possible to identify the topology of the magnetic fields of the outer part of the region, but there is strong evidence that the inner part is on closed magnetic field lines, which map to the day side auroral zone. The outer part of the transition region is a plasma depletion/magnetic field compression layer. The structure of the transition region is similar to that surrounding flux transfer events, which leads to the deduction that the plasma and field signatures of flux transfer events may be the result of displacement of the transition region earthward. Cases where the displacement is such that the field maximum of the depletion/compression region is encountered may well explain “crater” flux transfer event signatures.

Electron measurements are studied in the transition region where the magnetosheath merges with the magnetosphere. The outer part of the transition is found to be a depletion region, in which a reduction in the electron thermal pressure coincides with compression of the magnetic field. It is suggested that the compression is associated with the draping of the interplanetary field around the inner part of the region, which is a region of closed magnetic field lines. Implications for the topology of boundary layers throughout the magnetosphere are discussed.

Two theories of auroral electron acceleration are discussed. Section 1 examines the currently widely held view that the acceleration is an ordered process in a quasi-static electric field. It is suggested that, although there are many factors seeming to support this theory, the major qualifications and uncertainties that have been identified combine to cast serious doubt over its validity. Section 2 is devoted to a relatively new interpretation in terms of stochastic acceleration in turbulent electric fields. This second theory, which appears to account readily for most known features of the electron distribution function, is considered to provide a more promising approach to this central question in magnetospheric plasma physics.

Particle measurements show that electron acceleration on auroral field lines is a statistical, velocity-dependent, process. It is proposed that the process is stochastic acceleration by waves, and demonstrated that lower hybrid waves seen on auroral field lines have the right properties to account for the electron acceleration. It is further shown that the lower hybrid wave power measured on auroral field lines can be generated by the streaming ions observed at the boundary of the plasma sheet, and that this wave power is sufficient to account for the electron power observed close to the atmosphere.

The interaction of comets with the solar wind depends on the ionization of the heavy cometary neutrals (mostly H2O and its dissociation products O, OH) which flow out from the nucleus, and the coupling of these newly produced cometary ions with the solar wind through its embedded magnetic field. The 'pick-up' of these heavy ions slows the solar wind such that a shock may form. As expected, the structure of such a shock transition is highly complex because the gyroradius of a heavy cometary ion is much larger than that of a solar wind proton. Here we present a comparative study of the solar wind electrons and protons and the cometary pick-up ions measured by Giotto at the inbound crossing of the bow shock at comet Halley. We find a highly structured shock transition starting with a cometary ion 'foot' seen at a distance on the order of the cometary ion gyroradius upstream from a sharp decrease in the solar wind proton speed. The total solar wind thermal and magnetic pressure is dominated by the relatively small population of cometary ions throughout the shock region.

UKS observations of the ionized barium and disturbed natural plasmas 170 km from the centre of the artificial comet release of 27 December 1984 lasted over 4 minutes. During this disturbance, solar-wind ions were retarded and deflected southwards and dawnwards, while barium was accelerated northwards. Substantial electron energization was seen throughout, varying with time and with the energy of the source electrons. Rapid electron and ion density changes were observed in later stages

The decision to include a third spacecraft, the UKS, in the AMPTE mission was made in 1981. The reasons for this are presented, together with a description of the spacecraft, its subsystems, and a summary of its early orbit performance. The UKS scientific instruments, and early results from them, are described in companion papers in this issue.

A wave‐particle interaction model is proposed for the electron acceleration that leads to discrete aurora. Measurements of accelerated electrons are used to deduce the wave spectrum and the nature of the waves responsible for acceleration. The waves are identified as lower‐hybrid waves. Generation of the waves by various free energy sources is considered and the wave energy required to account for the aurora is calculated and compared with observations.

Pulsations in the intensities of electrons and positive ions observed on a rocket flight into a pulsating aurora are examined and discussed. The correspondence between electron and ion pulsations is investigated, and the velocity dispersion in electrons and ions is evaluated. A method is devised for identifying positive ions using a velocity spectrogram, though the results in this instance are inconclusive. Pitch-angle distributions of the electrons are found to be isotropic at both pulsation maxima and minima. Electron energy distributions are found to be approximately Maxwellian with temperatures, expressed in energy units E0(=kT), in the range 4–12 keV. Pulsations correspond to temperature changes of approximately a factor of 2.

The direct observation of high frequency ∼3 Hz modulation of electron intensities during pulsating aurora is comparatively recent although, from TV and photometric studies, modulation of the visual aurora at these frequencies has been recognized for over a decade. We report here 2.2 ± 0.5 Hz oscillations in the intensities of 4–25 keV electrons producing a pulsating aurora. The electrons were measured from a Petrel sounding rocket launched from Kiruna, Sweden on 25 January 1979. The oscillations like the slower 1–20 s pulsations, exhibit a marked velocity dispersion implying an equatorial origin for both forms of modulation. VLF—hiss observed at about the same time at the Satellite GEOS 2, operating in the same equatorial region, shows remarkable similarities to the modulation in the particle fluxes. A connection between VLF emissions and auroral pulsations has been suggested by Coriniti and Kennel who argue that low frequency, 5–300 s period, micropulsations modulate whistler-mode wave amplitudes leading to variations in the rate of pitch-angle scattering of electrons from the magnetosphere. The isotropic angular distributions reported here suggest that this mechanism acting alone cannot explain the modulation in the present case.

The characteristic energy of electrons producing pulsating aurora has been investigated by measuring the I(557.7)I(427.8) column emission ratio in auroral forms in the magnetic zenith. A decrease in the ratio is regularly seen at pulsation maxima but the change in deduced electron characteristic energy is not large. In one form the change was from 2.2 to 2.6 keV, while in another more active pulsating form the change was from 2.6 to 4. 0 keV. These are in good agreement with recent rocket measurements of electron spectra in pulsating aurora which show the spectral distributions to be Maxwellian with characteristic energies of only a few keV. This ratio technique appears to be of value in studying the morphology of pulsating aurora and the mechanism producing it.

Energy spectra of electrons encountered on a rocket flight across an array of auroral arcs are employed to test three related models of electron acceleration. All three are based on a potential difference existing between the source plasma in the magnetosphere and the observation point in the ionosphere. One of the models provides a satisfactory fit to the observed spectra. Two alternative mechanisms are suggested to explain this model. The first possibility is a time-varying potential difference, which results in the accelerated electrons being observed with a statistical distribution of energy gain. The second possibility, which results in the same energy gain distribution, is a constant potential difference operating in conjunction with plasma instabilities generated by the accelerated beam. The energy gain distribution in the second case is therefore a consequence of a constant potential difference and a variable energy loss. In addition it is suggested that electrostatic waves generated by the instabilities could accelerate ambient plasma to suprathermal energies. Application of the model to the complete data set yields a continuous record of the parameters defining the acceleration and source plasma across the array of arcs. Reference is also made to an acceleration mechanism involving resonance with electrostatic waves.