Statistical equilibrium calculations for the first ten levels of H and He+ were carried out for several model atmospheres of a star with Teff = 40,000 K and log g = 3.5, using the method of Mihalas and Kunasz (1978). Atomic level populations and line profiles were computed for “model C” calculated by Klein and Castor (1978) and for variants of this model in which material density and mass loss rates were scaled down by factors of 2.0, 2.2, 2.5 and 4.0. We will refer to these variant models as D, E, F, and G. The resulting line profiles are discussed. Two variants of the Klein-Castor model C are generally successful and produce He II λ4686 in emission with λ3204 in absorption. All normal Of stars observed with He II λ4686 in emission have λ3204 in absorption. In one of the models, Hα is in emission, while Hβ and the higher Balmer lines remain in absorption, as is commonly observed in Of spectra.

The Hα profile in the spectrum of Zeta Puppis is shown in Figure 1 (Conti and Frost 1977). It is of the P Cygni type, with a violet displaced absorption component that extends to about 300 km s−1 from line center. An unusual feature is the broad wing emission which can be seen out to 1500 km s−1 on both sides of the line. Of the roughly 2.5 Å equivalent width of Hα, 1.8 Å is due to velocity displacements |δv| > 300 km s−1. The presence of wing emission shortward of the violet absorption edge has been noted before in a Wolf-Rayet star, where it is attributed to non-coherent electron scattering (Castor, Smith and Van Blerkom 1970). In the present case, an optical depth in electrons of nearly 1.5 is required to produce the extensive wings by scattering a normal P Cygni profile. This exceeds by a factor of three the electron scattering optical thickness through the entire envelope, assuming published values of the mass loss rate, terminal velocity of the wind and photospheric radius. Thus an alternative explanation is required.

The CIII λ 1909 absorption feature (deepest at V=−1600 Km s−1) in this star, observed by Burton et al (1975) and Johnson (1978), was predicted by Castor and Nussbaumer (1972). It implies a copious flow of CIII ions, because the oscillator strength is only fik = 1.6× 10−7.

Several models for the winds of O stars have been proposed to explain the unexpected presence of high ionization potential ions such as N+4 and O+5. Lamers and Snow (1978) proposed that the winds of stars showing N V and O VI lines have elevated temperatures near 4±2 × 105K while cooler stars with anomalous Si IV lines have Te ≈ 7±3 × 104K. Alternately, Cassinelli and Olson (1978, CO) and Olson (1978) have explained the presence of these ions by showing that a thin corona at the base of a cool wind (Te ≲ Teff) can produce these ions by the Auger photoionization process where a single x-ray photon causes the ejection of two electrons. A third possibility is that the winds are at only slightly elevated temperatures (40 000 to 60 000K) and photoionization in an optically thick wind produces the unexpected ions. The problem of determining the local radiation field in such a wind is discussed by Klein and Castor (1978). The purpose of the present analysis is to test the ability of these three wind models to fit the observations of ζ Orionis A 09.7 Ib.

A great deal of work has been done on the theory of mass loss and evolution in close binaries, and numerous individual systems have been discussed in this connection, but the general question of the binary frequency of O-stars, and in particular, the initial binary mass ratio frequency or distribution of secondary masses, has not been completely answered. In general, we know that about half of all O-type stars are binaries; the most recent determination by Conti, Leep and Lorre (1977) found 58% of their sample to be certain or probable binaries. However, many of these stars were judged to be variable on the basis of only a few spectra from different sources, and therefore require further study. Another point to be examined concerns the binaries with available orbits: two thirds of these are double line systems. Figure 1 shows a plot of the semi-amplitude versus orbital period for all known systems, along with some theoretical curves for different mass ratios. Not only is the lack of single line systems obvious, but low amplitude systems are almost completely missing. This would appear to be only an observational selection effect, although it is to be noted that low amplitude double line Wolf-Rayet systems have been detected. If the effect is real, it implies that O-type binaries with mass ratios (m1/m2) greater than about three do not exist.

The eclipsing binaries UW CMa, AO Cas, and V729 Cyg have been systems of great interest for over fifty years. The light curves are complex and suffer significant changes on a time scale of months, but the primary attraction of these systems is that both components have O-type spectra; thus they present us with some of the few possibilities for direct measurement of absolute dimensions of very massive stars. Much effort has been expended on these systems, but no really consistent model has emerged.

HD 149404 (HR 6164) was noted to have double lines by Conti, Leep and Lorre (1977). The spectrum has been previously classified as 09Ia, and shows strong Si IV λ4089, 4116 absorption, He I stronger than He II, N III λ4634, 41 weakly in emission, He II λ4686 “filled in,” and all the Balmer lines through H13 in absorption except Hα, which is strongly in emission.

The star HD 93206 (=QZ Carinae) is a double-lined (Conti et al. 1977), eclipsing (Moffat and Seggewiss 1972) binary with a period of 6 d. Walborn (1973) classified it 09.7Ib:(n). Since the star is probably a member of the cluster Collander 228 (which is near η Carinae), its distance can be assumed to be 2600 pc. In principle, one can determine the masses of the components of HD 93206 from observations of the radial velocities and the light curve, and a spectroscopic orbit is the object of this investigation. A mass determination for an evolved star such as this one is especially important for checking recently computed evolutionary tracks with mass loss for massive stars (de Loore et al. 1977, Chiosi et al. 1978, Dearborn et al. 1978).

The far-UV spectrum of the eclipsing binary UW CMa (O7f + O-B) had earlier been utilized to derive a mass-loss rate of about 10−6 to 10−5 solar mass per year. The mass flow seems to be basically in the form of a stellar wind emanating from the O7f primary component, with radiation pressure as the controlling factor. The main characteristics that make UW CMa a possible progenitor of a Wolf-Rayet system are discussed.

γ2 Velorum is the brightest binary system with a Wolf-Rayet component; it has been classified as a 09I+WC8 pair (Conti and Smith 1972). The orbital elements of this system were determined earlier by Ganesh and Bappu (1967) from spectrograms of 125 Å/mm dispersion, on the basis of the radial velocities from CIII-IV λ4652 Å emission for the WC component, andon the radial velocities from H absorption only, for the O star.

HD 92740 is a star located in the Carina nebula showing a Wolf-Rayet spectrum of type WN7. Faint absorption lines of the upper Balmer series of hydrogen, and also of the Pickering series of Hell are present in the spectrum, in addition to the WN emissions. Although absorption lines present in a Wolf-Rayet spectrum are generally assumed to arise in a companion OB star, a previous study (Niemelä 1973) of the radial velocities of HD 92740 showed that the absorption and emission lines followed the same orbital motion. Subsequent spectral observations of this star have been carried out at the Cerro Tololo Inter-American Observatory, Chile, and at the Córdoba Observatory, Argentina, during four years; the observational data are listed in Conti, Niemelä and Walborn (1978). These observations showed that the true period is 8 times longer than the initially derived period of 10 days, and that doubtlessly the absorption lines belong to the WN star.

A coordinated campaign of radio, optical, and x-ray observations of the bright x-ray source Cygnus X-1 took place August 7–21, 1977 under sponsorship of Commission 42 of the International Astronomical Union. Radio flux measurements, optical spectra, photometry, and polarimetry were obtained during this period by ten groups from Canada, Great Britain, the Soviet Union, and the United States. The x-ray flux was monitored continuously by the SAS-3 satellite between August 11.0 and 17.0.

We propose that most of the OB runaway stars are Old Disk Population objects in the same evolutionary phase as the hot UV-bright stars in globular clusters. Bimodal Gaussian fits to the peculiar radial velocity distribution are computed for 386 O-type and 1093 B-type stars. Both samples independently yield one Gaussian with σ ≃ 13 km s−1, a value typical of extreme Population I objects, and a second one with σ ≃ 28 km s−1 which is characteristic of the Old Disk Population. The fraction of stars under the high velocity-dispersion distribution (HVD stars) is 47% of the 0 and 23% of the B stars. We analyze the kinematics of the sample of OB stars divided into low peculiar radial velocity, |Vrp|<20 km s−1, and high-velocity stars, |Vrp|> 45 km s−1. The results for the solar motion and mean peculiar velocity of the groups are the expected ones for the extreme Population I objects in the case of the low-velocity group (U⊙= 9.1±0.1;V⊙= 14.8±0.1 km s−1) and an asymmetric drift of approximately 20 km s−1 for the high velocity stars (U⊙=2.7±1.4; V⊙= 32.7±1.4 km s−1). This lag behind circular motion also corresponds to Old Disk Population objects.

We have obtained the observed fraction of supergiant (luminosity classes I and II), giant (III) and dwarf (IV-V) stars of spectral types B2 and earlier. The stellar sample used was formed with all the stars with bi-dimensional spectral classification listed in the Catalogue of Galactic O stars by Cruz-González et al. (1974), the unpublished compilation of B0 and B0.5 stars by J.F. Rayo, and the B1-B2 stars listed by Morgan et al. (1955). The latter sample is by far the least complete one. The results are listed in Table I, together with the total number of stars (in parenthesis) considered in each spectral interval. A prominent conclusion is drawn from the table: The fractions remain approximately constant all over the spectral range considered.

Some conclusions concerning the luminous hydrogen-deficient stars are presented, as is also a list of 45 such objects.

1. A. Mass loss has been observed by means of UV, optical, IR and radio observations.

2. B. There exists also indirect evidence that stars lose matter during their evolution.

Although it was known since a long time that very luminous blue and red stars may show evidence of mass outflow, it was the advent of the Copernicus satellite that clearly ascertained all OB supergiants (and also the most luminous main sequence stars) are losing mass at rates that are significant for their evolutionary history. The observational information for blue luminous stars (O, Of and WR) has been recently reviewed by Conti (1978), who discussed in some detail the data available for different spectral regions. The rates of mass loss inferred from these observations are estimated to be in the range 10−7 to 10−5 M⊙/yr for OB stars, and from 10−5 to 10−4 M⊙/yr for WR stars. Several theoretical models have been proposed to explain those high mass-loss rates, and the high terminal velocities ranging from 1000 to 3000 km/sec. The two basic models are: the cool radiation pressure model, originally proposed by Lucy and Solomon (1970) and elaborated by Castor et al. (1975), in which the envelope is accelerated by momentum transfer from radiation to ions due to the ultraviolet line absorption; and the coronal model of Thomas (1973), and Hearn (1975), where the wind is sustained by gas pressure in a hot corona around the star. As both models do not completely account for the observations, several complementary modifications have been suggested. The nowaday situation is reviewed by Cassinelli and Lamers (1978), and Conti (1978).

Many of the effects of mass loss on OB stars have now been explored. Mass loss will cause a star to be overluminous for its mass (though less luminous than a star of its original mass) and, for moderate mass-loss rates, the luminosity decreases at the same rate as the mass contained in the convective core decreases causing the main sequence lifetime to remain unchanged (Chiosi and Nasi 1974, 1978, Deloore, DeGreve and Lamers 1977, Dearborn, Blake, Hainebach and Schramm 1978). Mass loss can also expose layers where 14N has been enhanced via the CNO tricyle (Dearborn and Eggleton 1977) and, in extreme cases, can produce a stripped helium core resembling a Wolf-Rayet Star (Hartwick 1967). While many of these phenomena (in particular the composition change) are more sensitive to the total mass removed than the formalism used to represent the mass loss, significant differences will result for the same average mass-loss rate depending on whether the mass was removed early (near the ZAMS), or late (near core hydrogen depletion). In addition, there appears to be a critical mass loss rate which depends on initial mass and separates those models which continue to evolve in a relatively normal (though subluminous) manner, and those models which evolve to a Wolf-Rayet configuration.

Observations of the chemical abundances in young supernova remnants may be used, in some circumstances, to place constraints on the evolution of the progenitor stars. For example, if a progenitor was massive (M > 10 M⊙), the presence of high 14N/1H ratios (that is, more than five time the solar value) in the supernova remnant can imply that substantial mass loss took place during the star's early evolution, that is, while it was an early-type supergiant.