We have taken advantage of the high spatial resolution attainable with the HST to map the linear polarization in the V band across the nebulosity surrounding Eta Car. There are several new results related to polarization variations on different size scales. First, we present a two-dimensional map of the amount and position angle of the polarization across the Homunculus. Second, we provide measurements of the polarization within prominent features such as the “jet”, the “paddle”, the “skirt”, and the “spot” in the south-eastern lobe. Third, we comment on polarization variations associated with the small-scale structure that can be seen in HST images (and which gives the lobes their cauliflower-like appearance). The new data provide insight into the three-dimensional distribution of dust about Eta Car.

The trajectories of wind particles emitted from the surface of a massive star in a close binary system are analyzed. Within a radius 1000 times the separation of the two stars, a significant fraction of the particle trajectories are found to cross the orbital plane, leading to the formation of a large-scale, outflowing circumbinary disk-like structure. The shocks which arise due to the collision of particles which are within crossing streamlines are expected to produce selected regions within the wind, particularly in the orbital plane, in which a higher degree of ionization prevails than in the wind in general. X-ray emission might also be expected from these regions. Such a model is suggested to be applicable to the erupting WR/LBV binary system HD 5980 and other binary LBV systems, during phases when wind velocities are slow.

P Cygni has been for a long time an often-used target of observations, and a test object to prove several modelling approaches (cf. e.g. Najarro et al. 1997, and references therein). However, the origin of it’s stable variability pattern with moderate amplitudes is poorly known yet. In this contribution I will concentrate on the spectroscopic variability. A more elaborated version of this paper considering many kinds of variability data on P Cygni will be published elsewhere (Kolka 1998).

The striking phenomenon in the spectra of P Cygni is the cyclic drift of subsequent individual components through the absorption part of the P Cyg-type profile. This is demonstrated in Fig.1a in the case of three Balmer lines in 1982 (the data adopted from Markova, Kolka 1989). Every new drift-cycle starts when the component with smaller Doppler-shift becomes deeper than it’s neighbour.

We present the results of high-resolution spectroscopic observations of two high-luminosity stars, MWC 314 and AS 314, obtained at the 6-meter telescope of the Russian Academy of Sciences. Both stars are suspected to be candidate LBVs in quiescence.

The formation and survival of dust around stars requires a physical environment very different from that believed to hold anywhere in a Wolf-Rayet stellar wind. The observed facts of dust formation by Wolf-Rayet stars force the conclusion that their winds are not homogeneous. They also allow us to deduce the types of inhomogeneity, including clumps and large-scale high-density wakes produced in colliding-wind binaries, that allow the formation of dust.

We present ISO-SWS spectra of B[e] stars. We find a wide diversity of spectral characteristics of B[e] stars, suggesting different origins for the circumstellar matter. Most B[e] supergiants show hot dust with weak amorphous silicate emission. MWC 300 has a warm dust shell with strong crystalline silicate emission; its evolutionary status is unclear.

The line-driven instability may cause pronounced structure in winds of hot, luminous stars, e.g., fragments of dense shells, strong reverse shocks, and fast cloudlets. We discuss the linear stability theory, including the line-drag effect, phase reversal due to the diffuse radiation field, and the relevance of so-called Abbott waves. Recent hydrodynamic simulations focuss on the influence of a time-dependent source function on the flow structure, and on the X-ray emission from wind shocks and cloud collisions.

I review simulations of Co-rotating Interaction Regions (CIRs) in line-driven stellar winds. Previous CIR models have been based on a local, Sobolev treatment of the line-force, which effectively suppresses the strong, small-scale instability intrinsic to line-driving. Here I describe a new “3-ray-aligned-grid” method for computing the nonlocal, smooth-source-function line-force in 2D models that do include this line-driven instability. Preliminary results indicate that key overall features of large-scale CIRs can be quite similar in both Sobolev and non-Sobolev treatments, if the level of instability-generated wind structure is not too great. However, in certain models wherein the unstable self-excitation of wind variability penetrates back to the wind base, the stochastic, small-scale structure can become so dominant that it effectively disrupts any large-scale, CIR pattern.

O stars are located in a domain of the HRD where nonradial pulsations are expected. Photometric surveys did not reveal pulsating O stars, showing that the amplitudes must be very small. Intensive spectroscopic studies yielded pulsation modes for very few O stars only, although many are line-profile variables. Wind contamination of many spectral lines is a major difficulty. Future concentrated spectroscopic efforts will undoubtedly increase the number of pulsating O stars.

The current status of our knowledge of pulsations in these stars is reviewed. From one specific example a critical attitude emerges towards the quantitative results reached so far. We also address the question whether non-radial pulsation can be the cause of the non-spherical time-dependent winds of these stars.

If observing time and number of photons are not the limit, it will probably be very difficult to find any Be star or BA supergiant that is not variable. Moreover, there is hardly any major set of observations that is not tempting to explain at least partly in terms of nonradial (g-mode) pulsations. Since a few years ago, such conjectures are also theoretically permissible because improved opacity calculations have established the classical к-mechanism as a viable source of pulsation driving (cf. Pamyatnykh, these proceedings).

Contrary to Be stars, it can for any given BA supergiant nevertheless be arbitrarily difficult to diagnose nonradial pulsations (NRP’s) with certainty because they need to be detected against considerable background ‘noise’ of other physical processes, most of which are related to mass loss and/or rotation. To make things worse, there is some evidence that NRP’s can have some effect on the dynamics of the mass loss. On the other hand, variable and non-spherical winds is the subject of this Colloquium, and this paper is accordingly biased towards the interplay between pulsation and mass loss.

Basic physical mechanism of the pulsational instability in OB stars is shortly reviewed. Similarities and differences in the driving effect acting in these objects and in the classical pulsating variables are emphasized. Updated theoretical instability domains in the H-R diagram, in the log g – log Teff and in the Period – log Teff diagrams are presented and compared with the observational data on β Cephei, SPB and some other types of hot variables.

We discuss a possible role of non-radial oscillations as a cause of mass-loss in hot stars. In particular, we propose a working model for the episodic mass-loss in Be stars. In this model, equatorial mass loss is thought to be driven by wave-breaking phenomenon of large-amplitude non-radial waves and a circumstellar disk could thus be formed around the equatorial plane of a rapidly rotating star. A kind of relaxation-oscillation cycle could be established between the Be phase and non-Be phase, in which an interplay between non-radial oscillations in stellar atmosphere and the circumstellar disk is essential. We also discuss a viscous decretion-disk model for the circumstellar envelope around Be stars.

Many physical factors, including radial and nonradial pulsation, rotation, radiation pressure, convection, magnetic fields, or dynamical instabilities may play important roles in the hydrodynamics of Luminous Blue Variables. We review the current status of hydrodynamic modeling of LBV envelopes, and describe results of our models using the one-dimensional nonlinear hydrodynamics code of Ostlie and Cox. We find that the models pulsate in several simultaneous radial modes, driven by the helium and Fe ionization к effect. The pulsations have quasi-periods between 5 and 80 days, with radial velocity amplitudes of 50-200 km/sec, and may be identified with the LBV microvariations. In some cases, depending on luminosity-to-mass ratio and helium abundance, deep layers in the model can periodically exceed the Eddington luminosity limit. The key to exceeding LE is the inclusion of the time dependence of convection: Near the regions of opacity peaks produced by Fe and helium ionization, convection is turning on and off during each pulsation cycle. If convection cannot turn on rapidly enough to transport the required luminosity through the region, the Eddington limit is exceeded. If this region of the star is sufficiently adiabatic, an “outburst” may occur. In the hydrodynamic models, an outburst is indicated by the photospheric radial velocity suddenly becoming very large, and the photospheric radius increasing monotonically over several pulsation cycles. Such pulsation-triggered outbursts may be responsible for the driving of variable, nonspherical winds. If large and infrequent enough, these outbursts may be identified with the classic LBV eruptions accompanied by episodic mass loss.

The occurrence and properties of strange modes and associated instabilities in massive stars are reviewed. If applicable, strange modes may be classified by the maximum of the opacity they are associated with. Whether they are still present in the limit of vanishing or infinite thermal timescales – which corresponds to the NAR and adiabatic approximations respectively – is another criterion for classification. A model for the instability mechanism of strange mode instabilities is discussed.

Using a dynamical stellar code, detailed evolutionary models have been computed that follow the evolution and nucleosynthesis of massive stars from the pre-MS (accretion phase) up to the onset of O-burning. Various instabilities occur during the evolution. As an example, the evolution of a star with maximum mass 120 Mʘ is discussed.

We describe present theoretical ideas about the time evolution of the winds of luminous stars with emphasis to effects of non-sphericity and non-stationarity. We discuss the evolution of the winds of rotating luminous stars during their main sequence evolution, in particular when they approach their Eddington-limit or any other surface instability. We then consider the winds of post-main sequence stars up to the immediate pre-supernova stage. We connect the giant outbursts of Luminous Blue Variables with luminous rotating post-main sequence stars in thermal disequilibrium. We further discuss the spin-up effect of Heger & Langer (1998) for post-red supergiants and describe its observational consequences. We compare theoretical models with observations of the winds of B[e] supergiants and Luminous Blue Variables in general, and with SN 1987A, VY CMa and η Car and the Pistol Star in particular.

The expressions of the radiative flux at the surface of a non-uniformly rotating star are revised and this leads to a small extra-term in the von Zeipel theorem. The Eddington factor needs also to be carefully defined in a rotating star, as well as the critical break-up velocity. This leads us to reconsider the so-called Ω–limit. The most massive stars reach the Ω and Г limits almost simultaneously.

We also examine the latitudinal dependence of the mass loss rates Ṁ(υ) in rotating stars and find two main effects: 1) the “geff” effect which enhances the polar ejection; 2) the “opacity effect” (or “к–effect”), which favours equatorial ejection. In O–stars, the geff effect is expected to largely dominate. In B– and later type stars the opacity effect should favour equatorial ejection and the formation of equatorial rings. Possible relations with η Carinae and the inner and outer rings of SN 1987 A are mentioned. Opacity peaks produce some extrema in Ṁ(υ) and this may also lead to the formation non-equatorial symmetrical rings.

Anisotropic stellar winds remove selectively the angular momentum. For example, winds passing through polar caps in O–stars remove very little angular momentum, an excess of angular momentum is retained and rapidly redistributed by horizontal turbulence. These excesses may lead some Wolf–Rayet stars, those resulting directly from O–stars, to be fast spinning objects, while we predict that the WR–stars which have passed through the red supergiant phase will have lower rotation velocities on the average. We also show how anisotropic ejection can be treated in numerical models by properly modifying the outer boundary conditions for the transport of angular momentum.

New results for 60 M⊙ stellar models with different rotation rates are presented. Some effects of rotation on Wolf-Rayet star formation are discussed.

We present spectra obtained with the Infared Space Observatory (ISO) of the dust shells surrounding several Luminous Blue Variables (LBVs), both in our galaxy and in the LMC. The 20-45 μm spectra of R71, AG Car and Wra 751 show prominent emission features from crystalline silicates. The composition of the crystalline silicates in LBV dust shells is compared to that found in other types of objects, such as (post)-AGB stars and red supergiants (RSG). Both Wra 751 and AG Car have a high ratio of pyroxenes to olivines. This suggests that the grains in both stars experienced very similar processing, but that this processing has lead to a higher abundance of pyroxenes compared to RSG. The dust composition of the three LBVs discussed here suggests dust formation in a cool outflow not unlike those of RSG.

The physical basis for interpreting observations of nebular morphology around massive stars in terms of the evolution of the central stars is reviewed, and examples are discussed, including NGC 6888, OMC-1, and η Carinae.