I review the properties of the X-ray source Cygnus X-3, and the evidence that is available that indicates that its companion is a Wolf-Rayet star. I focus on the interpretation of spectra of its infrared counterpart. Considering all data, I conclude that the companion is indeed a WR star, and I discuss the ramifications of this conclusion for our understanding of WR stars in general.

The existence of Wolf-Rayet binary systems containing a collapsed companion is predicted on evolutionary grounds. However, because of the absence of strong X-ray emission and the difficulty in obtaining consistent periodicities from the observed variations, none of the “classical” WR stars has been shown to be a WR+cc system. A discussion is presented of the line-profile variations which are predicted to occur on orbital timescales due to the interaction of the WR wind with the cc. We note the importance of numerical simulations of the flow structure of the WR wind in the vicinity of a collapsed companion for deriving diagnostic tools for the presence of a cc in a WR system.

We present preliminary MERLIN radio maps at 1.7 and 5GHz of the Wolf-Rayet star WR147 (AS431). These provide the highest spatial resolution images of this object at these frequencies to date. Our results confirm that WR147 comprises two primary radio sources lying on a N-S axis. Combining results at 1.7 and 5 GHz gives a spectrum for the southerly source consistent with optically thick thermal emission. The northerly source appears to be non-thermal. In addition, extended emission is observed at both frequencies which links the two sources. The northerly source is elongated in a direction roughly perpendicular to the axis of this extended emission, and the southerly source appears double on the 5 GHz map.

Strong transient chaotic high-frequency activity has been detected in WR148 on a time scale of a few minutes. The main property is the variability of the fluctuation power, not the light-curve. The same is true for WR136 but with a more permanent pattern. The power of fluctuations is of order of thousandth relative to the steady intensity. The maximum fluctuation power lies in the frequency band of 25–50 Hz.

A survey of bright O- and B-type stars has been conducted from the ROSAT All Sky Survey data, aiming at the detection of compact companions expected for Oe and Be stars. Concurrent Hα spectroscopy has been used to establish the level of activity of these stars around the time of the X-ray observations. The detection rate of OB and OeBe stars as well as their average X-ray over bolometric luminosities suggest little difference between these two groups of objects. Instead of accretion onto a compact object (NS, WD), the X-ray emission from OeBe stars may be produced by stellar winds as has been proposed for O- and early B-type stars in general. Flexibilities in the available models for X-ray emission from accreting WDs may allow the X-ray emission from such objects to remain below a detectable level. These investigations are now being extended to pointed observations in the ROSAT Archive.

The radio emission from WR125 (WC7+O9) has faded to a new low, apparently below its stellar wind level, while continuous, variable non-thermal emission has been observed from WR146 (WC6) and WR147 (WN8).

During the 1992–1993 observing season, WR3, 6, 16, 40, 66, 82 and 134 were monitored in fast photometry mode with time-resolution 0.005—0.01 s. Only WR6 reveals a possible period of P = 0.11 s (semi-amplitude A = 0.025 mag), which is close to the derived equilibrium period of a new-born pulsar in a binary system after the rapid phase of Roche Lobe Over-Flow from the original secondary component.

A personal, but critical summary of the papers at this conference is presented.