As the first result of spectroscopic monitoring of WR46 (HD 104994, WN3p) in 1991, a preliminary radial-velocity curve is presented with P = 6.8 hr, confirming the binary nature suggested from photometry.

Spectral observations of the WN3p star WR46 (HD 104994) obtained during June 1993 and January/February 1994 display large amplitude radial velocity variations of the strong emission lines Nv 4603-19Å and HeII 4686A, on a time scale of a fraction of a day. The most probable period found is 0.311 d, similar to the photometric period found by previous authors. The amplitude of the radial velocity variations of Nv emission is almost twice that of HeII. Noting the similarity of WR46 with low mass X-ray binaries, we suggest that the emission line spectrum corresponds to that of a luminous accretion disk in an evolved binary system.

Analysis of radial velocities of the WR binary WR133 yield semi-amplitudes KWR = 49±2 km s−1 and Ko = 21±2 kms−1 respectively and a mass ratio q = 0.43. The HeII 4686Â profiles show increased blue emission when the base of the wind-interaction cone is directed towards the observer.

Phase-resolved observations of the soft X-ray flux from the WN5+O6 binary V444 Cyg (WR139) have been obtained by ROSAT. The objective of these observations were: (1) to compare our results with previous observations made with the IPC (Moffat et at. 1982); (2) to explore the phase dependence of the X-ray variations; and (3) to look for evidence of the generation of X-rays from colliding winds in the system. ROSAT observed V444 Cygni with the PSPC five separate times, for a total integration time of about 20 ksec. Figure 1 shows the phase-binned light-curve along with archival IPC data. The observed X-ray variability can be modeled as a wind eclipse of an X-ray source by the WR wind. If most of the X-rays from the system are produced in a shock-heated region of wind interaction between the two stars, then the shocked gas has a large physical extent (r ≈ 140RW R) if the observed variability is produced solely by the eclipse of the region by the WR wind. However, since the O star makes a significant contribution to the total X-ray flux from the system, the actual size of the shocked interaction region is probably much smaller.

The bright star Sk—67°18 (Brey 5) in the Large Magellanic Cloud (LMC) contains an eclipsing binary system. Our radial velocity study reveals that the orbital period is almost exactly two days. The spectra also show that the star's primary component is not of spectral type WN, but that the star is rather an Of+O type binary where the primary is probably of type O3f*. Furthermore, Sk—67°18 appears to be a high-mass multiple system.

Optical spectroscopic observations of the eclipsing binary WR star HD 5980 in the SMC, obtained during different epochs, show a remarkable change of the WR type spectrum from spectral type WN3-4 to WN7-8. Radial velocities of the WR emission lines from spectra obtained in August 1992, indicate an orbital motion corresponding to the secondary component which is in front of the system during the primary eclipse. We compare the RV curves and phase-dependent line-width variations observed in 1992 with previously published data.

Five-colour polarimetry of a sample of twelve southern WR stars is presented. We suggest a method for separation the interstellar and intrinsic polarization components. Based on the observed data, a dependence of λmax on the terminal velocity v∞ is suggested. Its possible origin could be related to circumstellar envelopes.

Intensive broad band observations of WR stars by Moffat et al. reveals photometric and polarimetric variability on time-scales of hours, similar to that seen on narrow spectral features. The r.m.s. fractional photometric variation is found to be about 20 times the r.m.s. polarimetric variation whilst simple scattering theory predicts ratios of order unity. The possibility that this discrepancy is due to polarimetric cancellation of the Stokes parameters of light scattered from many dense blobs at random positions (known from spectrometric data to be present) is investigated analytically and by simulations. It is found that factors of 20 reduction in the polarimetric variability cannot be achieved for reasonable numbers of blobs and it is concluded that further polarimetric suppression by substantial blob optical depth and/or intrinsic blob emission must play a role. These results put further constraints on physical conditions in WR wind inhomogeneities.

We present some results on circumstellar envelopes using a Monte Carlo code we are developing. In contrast to analytical approaches, the effects of multiple scattering and the finite sizes of sources and envelope have been incorporated in the models. As an illustration, we fit the light-curve of V444 Cyg as well as the modulation of the linear polarization throughout its orbit.

New spectrophotometric, photometric and polarimetric observations of V444 Cygni confirm the basic conclusion that the WN5 star has a small core radius (rc < 4 R⊙) and a high core temperature (Tc > 60 000 K), which are characteristic of massive helium stars. Values of rc < 3 — 6 R⊙ and Tc > 70 000 — 90 000 K for the core of the WN7 star in the Cygnus X-3 system agree well with this conclusion. A clumping structure of WR winds is suggested. X-ray observations of colliding winds in WR+O binaries suggest radial expansion and anomalous chemical composition of WR winds.

High-mass X-ray binaries (HMXBs) represent an important stage in the evolution of massive binary systems. The compact object (in most cases an X-ray pulsar) not only provides information on the orbital and stellar parameters, but also probes the stellar wind of the massive companion, an OB supergiant or Be star. The X-ray luminosity directly depends on the density and the velocity of the wind at the orbit of the X-ray source. Important constraints on the stellar-wind structure can be set by studying the orbital modulation of UV P-Cygni profiles. In this paper different aspects of the interactive wind-accretion process are highlighted, such as the highly variable X-ray luminosity, the influence of the X-rays on the radiative acceleration of the wind inside the ionization zone, and the large-scale structures that trail the X-ray source in its orbit.

WR stars obey M-L-·-Teff-R-chemistry relations, which are in general model independent (with the exception of chemistry). The processes of WR formation (stellar winds, hydrodynamical mixing, binary mass transfer etc.) influence, however, the domain of parameters occupied by WR stars. We specifically examine the distribution of luminosities and H-contents of WN stars, which both support heavy mass loss rates, or possibly mixing, in the main sequence and WNL phases. Detailed studies of the number ratios WR/O, WC/WN, WC/WR, etc. are made for galaxies at various metallicities Z which exhibit extremely different WR populations. Good agreement of models and observations is found. It is also shown that the data are better explained if a certain fraction Φ (less than 10%) of the O-stars become WR stars, preferentially of type WNE as a result of Roche lobe overflow (RLOF) in binaries. This result necessarily implies that the fraction of WR stars, owing their existence to RLOF, is variable with Z, being nearly 100% at low Z and much smaller at high Z. We also identify several hydrodynamical developments physically required in stellar models. Among them, we collect the available observational and theoretical arguments supporting an important role of mixing in massive star evolution.

The influence of mass transfer in an OB-type binary on the colliding wind characteristics of the following phase of WR binary is examined. We look to the changes in composition and orbital characteristics. Special attention is given to the formation of WR+OB systems and the colliding wind characteristics. The mass-luminosity behaviour after mass transfer is examined. We discuss the influence of the internal state, the composition of the wind, and rotational effects on the mass gainer. Recent models of both single stars and close binaries allow us to derive the dependence of these characteristics on the initial masses of the components. From this, constraints for the wind regimes at the WR+OB stage result.

We present a first step towards complete stellar models for massive stars that treat the entire star including its wind. Our models are based on state-of-the art non-LTE atmospheric and interior modelling. The models yield the stellar evolution under the constraint of realistic atmospheres as well as line-blanketed continuum fluxes and H and He line profiles. We follow the evolution from OB to LBV, Ofpe/WN, and WR stars. We investigate the effect of line-blanketing during MS evolution. We adress the effect of an the extended atmosphere on the evolution and discuss the first results for WR stars.

We present new models of massive close binary evolution, using an up-to-date hydrodynamic binary code which includes time dependent mixing of chemical species in the stellar interior, and nuclear reaction networks for the CNO cycles as well as for the NeNa and MgAl cycles. Some of our models, tailored to fit γ Vel, provide for the first time detailed results for the production and ejection of 26A1 into the interstellar medium by WR binaries. From our results one can estimate the contribution of WR stars to the amount of 26Al presently seen in the Milky Way by γ-ray satellites.

It has frequently been suggested that Wolf-Rayet stars are present in symbiotic systems, in particular in symbiotic novae. We dismiss that suggestion as far as it concerns Wolf-Rayet of Population I. There may, however, be similarities to WR stars found as central stars of planetary nebulae. — When treating wind phenomena in our binary configurations, there are common properties which justify that for certain investigations Wolf-Rayet and symbiotic systems are considered together.

We present here starburst models based on the most recent grids of stellar evolutionary tracks obtained by the Geneva group. These new models, computed with enhanced mass loss rates during the main sequence and the Wolf-Rayet WNL phases, very well reproduce the luminosities, surface abundances and statistics of WR stars (Maeder & Meynet 1994). This change of the mass loss rates considerably affects the way the WR stars, born in a starburst's episode, are distributed among the different WR subtypes. We compare the theoretical predictions with recent observations and conclude that: (1) to reproduce the high observed ratios of WNL to O-type stars, a flat IMF seems to be required; and (2) the models which reproduce the best the observed characteristics of WR stars, i.e., those computed with an enhanced mass loss rate, can also account for the observed properties of the WR populations observed in starbursts. Moreover, the possible presence of numerous WC stars found in the low metallicity He2-10 A starburst by Vacca and Conti (1992), can only be accounted for when the high mass loss rate stellar models are used.

We present a scenario for the evolution of massive stars in which a new mixing mechanism (named global diffusion) is taken into account. This type of mixing stands on the critical Reynolds number and radiative viscosity (Schatzman 1977) and allows mixing of material to take place between the core and the surface during the whole evolution on a very slow time scale. The physical processes triggering global diffusion deserve further study. We find that stellar models of massive stars calculated with global diffusion offer interesting clues to understanding the properties of Wolf-Rayet stars and their location in the HRD.

The first results of new evolutionary calculations of massive stars are presented. The calculations cover ZAMS masses from 20—200 M⊙. Yields and abundances throughout the stars are presented.

In case of moderate matter mixing in the semi-convective zone the primary loses part of its mass on a nuclear time-scale. It looks like a CNO-supergiant during the first part of the helium burning stage. Another part of the mass is lost on the thermal time-scale. The primary looks like a WR star during the second part of the helium burning stage.