It is shown that the binary technique - a B star companion is used as a light source which probes the wind of the red giant primary - has yielded accurate mass-loss rates and wind velocities for 8 G to M (super) giants and (in some cases) estimates of wind temperature.

Eclipsing binary systems have in addition revealed that G and K supergiants possess extended chromospheres which could be detected outwards to ∼ 1 R* stellar radius) above the photospheres. Electron temperatures Te and hydrogen ionization ne/nH seem to increase with height up to at least 0.5 R* (ne/nH= 10−2, Te = 104 K at 0.5 R*), and the winds start to be accelerated at heights above ∼ 0.5 R*.

Mass-loss rates appear to increase steeper than linearly with L/g · R. It is shown that the observed mass-loss rates are consistent with stellar evolution constraints for both Pop. II and Pop I stars.

Results from 58 chromospheric eclipse spectra of three ζ Aurigae type binary systems - with a K supergiant primary and a B star companion -taken with IUE at high resolution are presented. Curves of growth have been constructed at 20 phases using selected chromospheric absorption lines superposed upon the B star spectrum. In order to fit average density models to the chromospheres, 3 samples of column densities (for ζ Aur., 32 Cyg. (fig.1) and 31 Cyg. K giants) have been used.

Observations made with the Infrared Astronomical Satellite (IRAS) showed the existence of a set of K stars with excess emission at 60 μ. The brightest star in the group is δ And (K3 III). Spectra of δ And have been obtained with the International Ultraviolet Explorer (IUE) in both the long and short wavelength regions. These spectra show several features unusual in a K giant as cool as δ And, in particular, emission from CIV, MgII h and k lines with blue wings stronger than the red and MgII absorption blue shifted to up to 300 km s−1. Overall δ And is similar to the known hybrid bright giant α TrA (K4 II). We have discovered the first ‘hybrid’ giant star. The IRAS observations are interpreted in terms of a cool (∼100 K) dust cloud surrounding δ And - a spectroscopic binary system- and a third component at 1200 A.U.

It is shown how empirically derived constraints affect models of the outer atmospheres and winds of K giants, taking α Boo (K 2III) as an example. The importance of empirical approaches prior to making semi-empirical models is stressed. The reliability of recent wind models is assessed.

Mass loss generated by radiatively damped acoustic waves is investigated. We find that a persistent wave energy flux leads to extended chromospheres. Mass loss is quite likely produced if the wave field retains a transient character and if large wave periods are used.

Infrared vibration-rotation bands of CO are ideal probes of the circumstellar environment of yellow supergiants. Results for a sample of stars are reviewed.

A bright enigmatic supergiant, 89 Her (HD 163506; F2Ia, Mbol=-7m.5), exhibits the light variations (Percy et al. 1979, Fernie 1981, 1983) as well as the changes of radial velocities and structures of Balmer and sodium lines. Sargent and Osmer (1969) have given some evidences for the existence of expanding circumstellar envelope around 89 Her and have discovered 24 emission lines of neutral metals in its spectra.

We present an overview of recent stellar X-ray observations, with some discussion of the requirements for future observations. We argue that solar observations indicate that coronal X-ray emission is strongly related to surface magnetic field activity; we show that the interpretation of X-ray stellar coronal emission from late-type stars within the framework of models analogous to those developed for the solar corona is viable, and it is supported by many experimental results. The extension of this solar analogy to the early-type stars is quite questionable and remains an unsolved problem, while the working hypothesis of an X-ray phase, related to phenomena of magnetic field-related activity, as contrasted to a wind phase during the PMS evolutionary stage is suggested by the present status of observations.

X-ray observations have shown in recent years that the existence of high temperature plasmas in stellar atmospheres is far more widespread than extrapolated from preexisting theories of the solar and stellar coronae, forcing a radical change in our understanding of the mechanisms of coronal physics. This paper reviews our current ideas on stellar coronae, in particular on the role of magnetic confinement in the atmospheres of late type stars.

We suggest that the conspicuous gaps frequently seen in the color-magnitude diagrams of globular clusters are caused by a sudden increase in chromospheric activity shortly after stars evolve away from the main sequence.

In order to determine the heating mechanisms for stellar transition regions and coronae we try to determine the damping lengths for the mechanical flux(es) responsible for the heating. For the lower part of the transition regions (30,000 < T≤100,000 K) the damping lengths are consistent with Shockwave damping. This appears to be also true for the upper part of the transition region in Procyon, while for the upper part of the solar transition region the damping length is much larger.

A star cannot have a solar-like corona if the available mechanical energy flux in the chromosphere is either too large or decreases outward more rapidly than the pressure. This result might be relevant for hybrid stars and cool giants.

The chromospheres and transition zones of the fast rotating giants of the FK Comae type can be studied by analysing their ultraviolet emission line spectra. From relative line intensities, electron densities of the order of 1010to 1011cm−1 are found for the region where the Si IV emission arises. The sizes of the chromospheres and transition regions can be inferred from the emission measure distribution, and a temperature-height relation can be found on the assumption that hydrostatic equilibrium holds. We find the atmospheres of these stars to be clearly more extended than those of normal giant stars, and the flux in the higher excitation chromospheric and transition zone lines (e.g. C II, C IV, Si IV) is significantly stronger than in other stars of similar spectral type. Indeed, the location of these stars in the standard rotation-activity-correlation diagrams places them close to or even above the saturation limit for main sequence stars.

We are currently working with a statistically complete, unbiased sample of 125 x-ray-bright stars which were serendipitously detected by the Einstein Observatory Medium Sensitivity Survey (MSS). A program of optical spectroscopy and photometry is currently underway to measure radial velocities, distances, and such stellar parameters as rotation, temperature, surface gravity, metallicity, chromospheric activity, and age and to correlate them with absolute x-ray luminosity. So far, the majority of the sample (which was defined at |bII| > 20o) appears to be composed of either flare stars (e.g. dMe, dKe) or active binary systems (e.g. cataclysmic variables, RS CVn, W UMa).

The study of the relationships between various activity diagnostics in T Tauri stars (TTS) suggests that the CaII K, MgII k and H∝ lines are formed in a similar region of TTS' atmosphere. For the more active TTS, an extended circumstellar region seems to be the major source of the emission, whereas a solar-type atmosphere alone may be able to account for the emission spectrum of low-activity TTS.

The Einstein Observatory demonstrated the existence of hot envelopes, i.e., stellar coronae, around most classes of normal stars (Vaiana et al. 1981). The coronae of late type stars of spectral type F through M are generally thought to be solar-like, i.e., structured and organised by the magnetic field topology and heated by some process(es) involving magnetic energy. Here the property “solar-like” does not refer to the optical appearance of a star, but rather to the role played by magnetic fields in the outer stellar envelope (Linsky 1985). Since it is difficult to measure magnetic fields on other stars directly, a number of indirect indicators is used in order to infer whether a corona should be considered “solar-like” or not.

Qualitative results on the H and K CaII lines in R Leo are presented from spectra throughout all the cycle of this mira star. A scenario of explaining the observations by an outward Shockwave is proposed.

Presence of CO layer well separated from photosphere is confirmed and this revealed a presence of quasi-static turbulent transition layer in normal red (super)giant stars. This layer may be related to an outer part of the extended chromosphere and/or a cool part of the chromospheric inhomogeneity, and will play major role in stellar mass-loss.

The IRAS catalogues have been searched for cool (G,K,M) giant and supergiant stars to investigate the occurrence of circumstellar (C/S)silicate dust, revealed by its emission features at 9.7 and 18μm. Low Resolution Spectrograph (LRS) spectra covering the 7-23μm range were used, plus the 60 and 100μm photometric points. M Supergiants were found in White Wing (1978), other stars by correlating the Bright Star Catalogue with the LRS catalogue: this discriminated against very cool stars reddened by dust; however i t can be seen in Table I that there is a clear trend for cooler and more luminous stars to have a dust shell. M Supergiants almost all have dust shells, whilst only the cooler M bright-giants and giants do. Of the G and K stars, only a very few of the Supergiants have dust shells. The sili cate features fell into two categories:

Evidence for circumstellar absorption around the warm N-type carbon star TX Piscium was found in a high-resolution IUE spectrum by Eriksson et al. (1986). This investigation also included the search, with a positive result, for CO J=1−0 emission from the circumstellar shell. From the Mn I absorption and CO emission a column density of about 1020–1022 H atoms per cm2 was estimated, as well as a mass loss rate around 10−7–10−6 M per year.