Since the recent discovery of fast pulsations of a WR star (Blecha et al. 1992), we observed five WR stars intensively and we report here the detection of fast photometric variations for three of them. These variations have a quasi periodic features with a time-scale compatible with theoretical predictions (Maeder 1985, Glatzel et al. 1993). For all observed WR stars the period can vary significantly and even disappear, showing the erratic nature of this phenomenon.

We present the results of a multifrequency observation of the WO star Sand 2 in the LMC. Two ROSAT PSPC observations, with exposure times of 8953 s and 4540 s, are used to investigate possible X-ray emission from Sand 2. The optical spectrum observed with the ESO 2.2m telescope close to the time of the ROSAT observations, shows strong emission lines of HeII, CIV, OV and OVI. All lines have a velocity broadening of 4 200±100 km s−1. The UV spectrum from the IUE archive shows a CIV resonance doublet at 155 nm with a P-Cygni absorption at ∼ −5 000 km s−1, with wings extending to about −8 000 km s−1. A combination of two different BB temperatures (80 000 K and 25 000 K) are required to fit the UV-optical continuum, dereddened by EB-V = 0.25.

The WO stars form a subgroup of the WR class characterized by a strong O VI emission at 382 nm (Sanduleak 1971; Barlow & Hummer 1982). In 1989–90 we observed intermediate resolution spectra of the four Sanduleak WO stars and of their surrounding nebulosities using the Bologna-Loiano 1.52m and ESO-La Silla 1.52m and 2.2m telescopes (e.g., Polcaro et al. 1991, 1992, 1994 in preparation). Fig. 1 shows that the spectra are similar but with some striking differences among the four stars. They can be arranged in a sequence in order of increasing ionization: Sand 2, Sand 1, Sand 5, Sand 4 (see Table 1). Sand 4 is a binary system, and the continuum of the O4 companion dominates over that of the WO star. In the other objects, the reddening-corrected continuum can be fitted by a 80 000 K BB, in agreement with the high ionization level of their emission spectra.

We present 2D numerical simulations of evolving ring nebulae around moving Wolf-Rayet stars. We have considered the interaction of the fast WR wind with the circumstellar matter modified by the previous action of the slow RSG wind. The resulting shell morphology is far from spherical and shows scallops due to several kinds of instabilities. The exact appearance of the ring nebula depends on the value of several parameters such as the unperturbed medium density and the velocity of the star.

We follow the interaction of massive stars with their circumstellar media over their entire lifetimes by combining several numerical models. We use a stellar evolution model for 35 and 60 M⊙ stars (Langer et al. in preparation) as inner boundary conditions for one- and two-dimensional gas dynamical models of the circumstellar medium using ZEUS (Stone & Norman 1992). In this poster, we focus on circumstellar evolution during the Luminous Blue Variable (LBV) and Wolf-Rayet (WR) stages.

High-resolution IRAS images, processed by HIRES, are ideal for the search of large, cold, fossil shell structures around Wolf-Rayet stars. In our preliminary effort, we have confirmed the large, 1.°5 — 2°, ring structures around NGC 6888 and RCW 58, which are clearly shown in the HIRES processed images. We interpret them as fossil bubbles blown by the main-sequence progenitors of the WR stars, although these large shells were previously suggested to be evolved supernova remnants.

We have observed the southern HII region RCW 49 with the ROSAT PSPC instrument. Part of the diffuse X-ray and optical emission present in this region might be associated with the X-ray selected WR star Th35-42 (WR20c). The young star cluster Westerlund 2 (which contains WR20a) is seen in X-rays as a centrally peaked, resolved source, surrounded by fainter diffuse emission.

Coronagraphic and spectroscopic observations of the nebula M1-67 surrounding WR124 reveal an axisymmetric outflow.

We observed in the Large Magellanic Cloud the associations LH 101 and LH 104, containing six Wolf-Rayet stars: one WN and five WCE subtypes. These regions are located south of the 30 Dor nebula. The new observations include UBV photometry for a thousand stars and spectroscopy in the range from 3600 to 5300 Å for about 90 stars. In LH 101 an 06.5f, an 06.5 + WN8 and a hydrogen emission star are identified. Cluster HR diagrams of these OB associations reveal their age structure.

Preliminary results of our NGC 300 Wolf-Rayet star survey are presented.

A number of central stars of planetary nebulae (CSPN) show emission lines in their spectra. Among them about 50 have been classified as [Wolf-Rayet] type (Tylenda et al. 1993). In spite of the fact that these stars probably have masses below 1 M⊙ and luminosities < 104 L⊙ their spectra show emission bands very similar to those of much more massive and luminous Population I Wolf-Rayet stars. We present some preliminary results of an extensive investigation of the [WR]-type CSPN based on the spectroscopic survey described in the Strasbourg-ESO Catalogue of Galactic Planetary Nebulae (Acker et al. 1992).

Observational data suggest that WR atmospheres must be considered as highly inhomogeneous objects incorporating strong shocks induced by radiative instability and full-scale supersonic turbulence, with relatively small mean mass-loss rates.

The UV — IR properties of the continuous energy distribution of single WR stars in the Galaxy and LMC are reviewed. The observed, dereddened distributions may be approximated as a power law Fλ ∼ λ−α with a non-Rayleigh Jeans range of α directly associated with the emission of excess flux over the full wavelength range and across all spectral subtypes. The observational results are compared to published NLTE models, where we find an important discrepancy in the frequency distribution of α for Galactic WN stars. A dependency of emission line equivalent widths on the continuum flux at line center is discussed; the systematic offset of the absorption line (but otherwise presumed single) objects is examined, and the observed behavior is compared to models and explained with simple theory.

Stratified non-LTE models for expanding atmospheres have become available in the recent years. They are based on the idealized assumptions of spherical symmetry, stationarity and radiative equilibrium. The satisfactory agreement between calculated and observed Wolf-Rayet spectra suggests that this “standard model” is basically adequate for describing real WR atmospheres and hence can be applied for their quantitative spectral analyses. By the application of these models, the fundamental parameters have been determined meanwhile for the majority of the known Galactic WR stars. Most WN stars populate a vertical strip in the Hertzspung-Russell diagram at effective temperatures of ≈35 kK, the luminosities ranging from 104.5 to 105.9L⊙. Only three WN stars of earliest subtype, other early-type WN stars if they have strong lines, and the WC stars are hotter. The chemical compositions of the WR atmospheres correspond to nuclear-processed material (WN: hydrogen burning in the CNO cycle; WC: helium burning). Hydrogen is depleted but still detectable in the cooler members of the WN subclass. Quantitatively, the hydrogen abundances show an interesting correlation with the luminosity which can be compared with the predictions from evolutionary calculations.

In the last decade there has been a tremendous advance in our understanding of the atmospheres and stellar winds of Wolf-Rayet (WR) stars. This has arisen through improved observations, and through the application of extensive radiative transfer codes which has allowed us to model some of the complex processes occurring in WR envelopes. Present models are able to simultaneously treat detailed atomic models for H, He, C, N, O and Si. In the near future we will be able to treat Fe, and to more accurately treat the formidable problem of non-LTE line-blanketing. In this review we address current limitations in modeling the extended atmospheres of WR atmospheres. These limitations fall into two classes — those reflecting our inability to accurately compute the “Standard Model”, and those referring to assumptions in the “Standard Model” that need to be relaxed. Under the first category we include inaccuracies in the atomic data, the neglect of line-blanketing in the model calculations, and the inability of the present generation of models to compute the velocity law. Under the second category we include the assumption of spherical geometry, the assumption of homogeneity, the neglect of rotation, and the influence of X-rays arising from shocks in the stellar wind.

In this paper I examine the acceleration of Wolf-Rayet winds in the optically thin part of the atmosphere. First, I investigate the radiative force using atmosphere models with specified velocity structures. It is found that “standard” model atmospheres that reproduce the observed spectra of Wolf-Rayet stars do not provide by a large factor (≈10) the force needed to accelerate the adopted velocity structure. Second, using clumping in order to have a favorable luminosity to mass loss ratio, it was attempted to find a hydrodynamic solution of a WR wind. If large clumping factors of the order of 30 are assumed then solutions are possible. However, such models are very unlikely to be correct since they predict much stronger emission lines than observed in any WR star. Third, a modification of the non-LTE rate equations is investigated. It is assumed that a Bowen resonance-fluorescence mechanism removes a small fraction of photons from the radiation field of the helium Lyα resonance line. It turns out that the ionization equilibrium of winds where helium is recombining from He++ to He+ is extremely sensitive to a modification of the radiation field of this line. A removal of 1% of the photons is sufficient to initiate an abruptly changing ionization equilibrium. It is proposed that the key to understand the acceleration of WR winds is the correct calculation of the ionization structure. If this hypothesis is true then today's non-LTE atmospheres yield systematically wrong results. This would not only affect the radiative acceleration but also the observable helium lines that are used for the diagnostic of the stellar temperature. It would imply that the luminosities of WR stars are underestimated by the present atmosphere models.

We review the dynamics of Wolf-Rayet stellar winds, with emphasis on how multi-line scattering can lead to mass loss rates for which the wind's radial momentum flux Ṁv∞ greatly exceeds the limit for single-scattering of the star's radiative momentum L*/c. The geometrical considerations that allow this to occur are illuminated through connections to multiple scattering in an effectively gray medium. The so-called “WR wind momentum problem” would be better characterized as an “opacity problem” of simply identifying a sufficiently dense ensemble of optically thick lines.

We discuss the evolutionary and mass loss implications of recent non-LTE analyses of late WN (WNL) stars in the Galaxy, LMC and M33 using the Wolf-Rayet standard model and address the observed dichotomy of WNL stars discussed by Moffat (1989). Individual subtypes belong to two distinct groups. We find that the single WNL+abs and WN7 stars evolve directly from very massive O stars (60–100 M⊙) with the former intimately related to extreme Of stars and found exclusively in the youngest clusters in our Galaxy. Conversely, the observational properties and chemistries of WN8-10 stars suggest that they are descended from lower initial mass progenitors (∼25–60 M⊙). These stars are either dormant LBVs (= WN9–10) or at a phase immediately after this stage (= WN8) although a previous RSG phase cannot be excluded. Included in this subgroup are the LMC and M33 Ofpe/WN9 stars, re-classified as WN9–10 since they appear to be genuine WN stars.

An analysis of the UV, optical and IR spectra of the composite WN/WC star WR8 is presented using the WR standard model. All spectral features are consistent with formation in the same stellar wind. The stellar parameters are T*=49kK, log L/L⊙=5.1, log Ṁ = −4.2, v∞=1590 km s−1. The derived chemistry (C/He=0.025, C/N=3, C/O=4) is intermediate between normal WN and WC stars.

We present an analysis of high resolution UV (IUE) and optical (INT, AAT and ESO) spectroscopy of eight galactic WN and WC stars. We examine the correlation between ionization potential (IP) and the velocity structure of the stellar winds and find important correlations between line-width, IP and excitation potential (EP). These results provide potential constraints on the velocity law used in models of WR atmospheres.