I have been asked to review in broad terms the concept of Wolf-Rayet (W-R) phenomena, that is, to outline what we currently know about the properties of stars showing such phenomena and to indicate the directions in which future work is leading. I thought I would begin by listing the characteristics of W-R spectra to which probably all the participants at this Symposium will be able to agree. These can readily be adapted from Thomas (1968) who discussed them at the Boulder W-R Symposium. The characteristics of W-R spectra are as follows:

1. (1) They are primarily an emission line spectrum superimposed on a “hot” continuous spectrum.

2. (2) P Cygni absorption components are observed for some lines in some stars; a fact not realized in 1968 was that a very few W-R stars have intrinsic absorption lines (e.g. The Balmer series in HD 92740 — Niemela (1973).

3. (3) The emission lines represent a wide range of excitation and ionization. This level is often higher than indicated by simple modeling of the continuous spectrum.

4. (4) The emission lines are broad, corresponding to widths of hundreds to thousands of kilometers per second; widths often differ among various ions in the same star.

5. (5) The W-R stars can be divided into two subtypes: These are the WN types, in which the ions of nitrogen dominate, and the WC types, in which the ions of carbon (and oxygen) dominate. Both sub-types have strong lines of helium; in a few cases, hydrogen lines, in emission, are also observed (Underbill 1968).

Nussbaumer et al. (1981) (NSSW) have calculated temperatures of 15 Wolf-Rayet stars, which cover most subtypes in the WN and WC sequences. Some Wolf-Rayet stars are found to have a blackbody-like energy distribution and therefore their effective temperature can be well determined, others show deviation from the blackbody shape. This deviation is probably due to an extended continuum emitting region. In view of these later cases NSSW adopted a temperature which they derived from a Zanstra analysis of the He II recombination lines. According to their analysis there should be a continuum jump at 2050 å. New IUE observations of the WN 5 star HD 50896 however did not show any trace of such an absorption edge. This implies that the effective temperature of this WN 5 star and probably of all WR stars with a non blackbody energy distribution is not yet known accurately. Therefore we started a detailed analysis with a spherically symmetric atmosphere model. Though the model has not yet reached its final shape we have obtained as a first result that the location of HD 50896 in the HR-diagram is to the left of the ZAMS and that the Zanstra analysis is indeed not valid for this star.

On the basis of the most recent compilation of narrow-band photometry and absolute visual magnitudes of Wolf-Rayet stars, and adopting a normal interstellar extinction law in all directions, the galactic distribution of 132 of the 159 known galactic WR stars is presented and discussed.

The spiral structure is found to be more clearly pronounced than in earlier studies. Furthermore we find an indication of two spiral arms at r=4 and 6 kpc. There appears to be an asymmetry of the z-distribution of single stars with respect to galactic longitude.

The location of the WC8.5 and WC9 stars between 4.5 and 9 kpc from the galactic center is discussed in the context of Maeder's red supergiant to WR star scenario.

We have analyzed high-dispersion spectra of over 100 WN stars in the Galaxy and the Large Magellenic Cloud (LMC). Most of the spectra were obtained through 4-m telescopes (Conti, Leep, and Perry 1981). The measured strengths of the WN classification lines N III λλ4634-41, N IV λ4057, N V λ4604 (Smith 1968) for stars of WN subtypes 2.5–8 are shown in Figure 1. The assignments of WN subtypes are based on visual estimates of ratios of line strengths according to the scheme of Smith (1968), except that stars appearing to be earlier than WN3 are classified as WN2 or WN2.5 as explained by vander Hucht et al. (1981) and by Conti, Leep and Perry (1981).

The amount of continuous energy from Wolf-Rayet stars and the shape of the continuous spectrum from the ultraviolet to the near infrared correspond to effective temperatures in the range 25000 to 30000 K. The value of log g is of the order of 4.0±0.5. Thus the photospheres of Wolf-Rayet stars correspond to those of moderately hot stars. The line spectra of Wolf-Rayet stars, however, indicate that electron temperatures greater than 30000 K occur in the outer atmospheres or mantles of these stars. Here outflow is important.

With the assumption of planparallel hydrostatic atmospheres, used generally for the computation of evolutionary models, the radii of WR stars are seriously underestimated. The true atmospheres may be very extended, due to the effect of the stellar wind. Instead of these hydrostatic atmospheres we consider dynamical atmospheres adopting a velocity law. The equation of the optical depth is integrated outwards using the equation of continuity.

The “hydrostatic” radii are to be multiplied with a factor 2 to 8, and the effective temperatures with a factor 0.8 to 0.35 when Wolf Rayet characteristics for the wind are considered, and WR mass loss rates are used. With these corrections the effective temperatures of the theoretical models, which are helium burning Roche lobe overflow remnants, range between 30 000 K and 50 000 K. Effective temperatures calculated in the hydrostatic hypothesis can be as high as 150 000 K for helium burning RLOF-remnants with WR mass loss rates.

Results are presented of a semi-empirical investigation of the transformation relations between colors and magnitudes in Lindsey Smith's narrow band photometric system for WR stars and their corresponding parameters in the broad band UBV system. A preliminary application of these relations to the study of WR stars in open clusters indicates that the intrinsic colors of WR stars are slightly bluer than the values tabulated by Smith, and that single WN stars are less luminous than previously supposed.

Radio observations of the Wolf-Rayet star HD 19 3793 are presented. Most of the observations were made during a large outburst in the infrared. The radio observations are consistent with the suggestion that a circumstellar dust shell recently formed.

Allen et al. (1972) observed 40 WR stars in the |1.6µ| and |2.2µ| bands, and found evidence of interstellar dust emission in the WC9 stars Ve 2–45, AS320 and HD 313643. Hackwell et al. (1974) reported 2.3 to 23 µ photometry of 19 WR stars and concluded that the excess infrared radiation from the Wolf-Rayet stars (except for the WC9 stars) could be explained by free-free emission from a hot circumstellar shell. Gehrz and Hackwell (1974) found from 2.3 to 23 µ photometry that three out of four WC stars appear to be embedded in thick circumstellar dust (graphite) shells, and concluded that WC9 stars may form a distinct Wolf-Rayet class. Cohen et al. (1975) derived energy distributions of 23 Wolf-Rayet stars from 3µ-11µ scanner spectrophotometry and infrared photometry, and concluded that WN stars show only free-free emission whereas only WC stars show dust. The excesses in WC9 stars are interpreted as thermal emission by graphite grains.

The spectra of a variety of Wolf-Rayet stars have been observed with ∼1% spectral resolution in the 1.4–4.1μm region using UKIRT. Strong lines due to ions of helium and carbon are observed and their relative strengths discussed. The He I singlet at 2.058μm is anomalously strong relative to other He I lines in WC stars and is responsible for the difference in the (H-K) colours of WN and WC stars. Emission line corrections to H, K and L magnitudes of different types are discussed. The Sanduleak 0 VI star ST 3 shows very strong C IV lines like the WC5 stars but not the strong He I.

Effective recombination coefficients have recently been calculated for recombination lines of He I, He II and C IV (among other ions with up to three electrons) for densities and temperatures appropriate for Wolf-Rayet atmospheres. These have been applied to recently obtained infrared spectra of γ Vel in order to derive the He+/He+2 and C+4/He+ + He+2 ratios.

This review summarises current knowledge of the chemical composition of PopI WR stars, concentrating on work carried out in this area since the last IAU, No. 49, symposium devoted to this stellar class (Bappu & Sahade 1973). Earlier reviews of this topic are found in Gebbie & Thomas (1968). The dichotomy of the WR stars into the WN and WC sequences (Beals 1934) has generally been qualitatively interpreted as arising because of gross differences in the C and N abundances: WN stars which exhibit emission lines of predominantly He and N ions with little evidence for C, being inferred as C-poor objects, whilst WC stars, showing predominantly He and C lines and virtually no evidence for N being inferred as N-poor. In both sequences the visible spectra show little or no evidence for hydrogen. However, although the WR stars have been acknowledged as a class for over a century now, progress has been very slow in putting quantitative determinations of their physical and chemical properties on a firm basis, with the bulk of work in this area being conducted during the past decade. The chemical nature of the WR stars has always been a matter of considerable uncertainty, controversy and, quite often, passionate disagreement, arising from uncertainties in the interpretation of the, often ambiguous, observational material available, as well as from disagreements as to the reliability of the use of comparatively simple analytical models employed to date. Recent results strongly suggest that the WR stars are chemically evolved objects, with low H/He ratios and quite different C/N ratios in the WN and WC sequences, with some measure of agreement in these results with the chemistries predicted to arise at various stages of evolutionary theory for hot massive stars which, by one means or another, have shed much of their atmospheric material during their evolution. My purpose in this review is to summarise the investigations and results that lead to the above conclusions. §2 deals with an assessment of the atmospheric H/He ratio in both WN and WC stars: a parameter of fundamental importance in addressing their evolutionary status, as well as providing a base species with which to compare other derived chemical abundances. §3 briefly deals with the models generally employed and gives recent results for He, C and N abundances derived from both visible and UV line analyses. §4 summarises recent results from stellar evolutionary theory and in §5 compares these with those derived from observation, assessing the significance of these new results and their implications for the evolutionary status of the WR stars. Some areas for further advancement are identified.

We have observed 45 WN stars in our Galaxy and the LMC with the IUE satellite at a resolution of 6 å. Examination of the spectra reveals surprising differences in line strength, in many cases at a given spectral subtype. This is similar to the description of the visual spectra given by Leep (1982). One generalization that can be made is that the equivalent width of He II λ1640 is correlated with the H/He ratio measured by Perry and Conti (1982), but not with spectral type.

The chemical abundances of hydrogen, helium, carbon, nitrogen, and oxygen in the winds of W-R stars bears directly on the problem of their evolutionary stage. The commonly held view that W-R stars are highly evolved objects has been based upon the apparent deficiencies of H in their envelopes and the enhanced strengths of CNO lines in the two sequences WN and WC (e.g. Gamow 1943). The picture that emerges is one in which many W-R stars are in the helium burning phase with the WC types more evolved than the WN stars. for this reason it is important to estimate the H content of as many WR stars as possible.

We present new determinations of the C/He, N/He and C/N ratios in 6 WN and 4 WC stars, resulting from a Sobolev analysis of new IUE UV line measurements in these species. Through improved observations of more stars, the present work extends an earlier, UV-based, study carried out by Willis & Wilson (1978, WW), which gives details of the Escape Probability Modelling (EPM) utilised. Full details of the current IUE observations, measurements and modelling will appear in Monthly Notices: here we summarise the results. Table 1 lists the ten WR stars currently studied — all were observed at low resolution and the brightest eight at high resolution.

I will comment on 3 problems concerning WR stars:

1. a. are WR stars core hydrogen/core helium burning?

2. b. are WR stars H-poor or do they have normal abundances?

3. c. can we use B.C. for WR stars which are similar as for 0 stars?

Until now, all that we've known about the magnitudes and colors of Wolf-Rayet stars has been based on photoelectric photometry made with 50–150 A wide interference filters, selected to exclude as far as possible the stronger emission lines (Westerlund 1966; Smith 1968, Lundstrom and Stenholm 1979). This was clearly an improvement on the pioneering efforts of Pyper (1966), who obtained broad band photometry and attempted to correct for the presence of emission; nevertheless, with modern detectors it is possible to go one step further.