The early history of the studies on Be stars is reviewed. The importance of keeping in mind a basic model of a Be star is emphasized, and the binary star model is suggested as one serious possibility at least for some Be stars.

My task is to review the spectroscopy and photometry of Be stars. In the wide range of topics this covers, I will naturally be biased by my own interests and knowledge. I hope however at least to mention all forms of observed behaviour within these limitations, but note that as we have specific sessions on energy distributions, rotation, observations at unusual wavelengths (i.e., not from 3500 to 6700 Å), and polarisation, I shall deal less thoroughly, if at all, with these topics. I shall also attempt to incorporate the results of the speakers who follow in this session, in order to save duplication of individual introductions and allow the later presentations to concentrate more on the relevant discussions. I shall deal with nine aspects of Be stars behaviour, under the headings below.

Hα observations have been made of the stars brighter than mv = 7.5 and north of δ = −30° contained in the three classical Be-star catalogues of Merrill and Burwell. At least one observation of each star was made at each of the two epochs 1958/9 and 1970/2 at the Lick Observatory and at the Warner and Swasey Observatory respectively. The dispersions used were 88 Å mm−1 at Lick and 66 Å mm−1 and 18 Å mm−1 (with image tube) at Warner and Swasey.

Results were as follows: The total number of objects observed was 215. Of these, 32 were supergiants and will be discussed later. Another 16 stars were deemed abnormal in some manner (helium and/or forbidden emission, composite spectrum, markedly violet-displaced Hα emission component: an interesting group of five stars including the nebular variable AB Aurigae). Thirteen of the stars showed no emission at either epoch. Of the 154 presumably normal low-luminosity stars showing emission, 70 displayed shell structure in Hα at one or both epochs, though in over half of these cases the shell absorption was considered ‘weak’. Thirty of the 70 stars varied in their shell characteristics between the two epochs. The total number of stars whose Hα line varied in some respect was 43, or 26 percent of the 167 normal low-luminosity stars observed. Quite a number of the stars observed have been reported to show shell characteristics by others but were not so noted by us.

Eight additional Be stars not contained in the Merrill-Burwell catalogues have also been observed. Seven of these are shell stars; the other is the unusually high latitude object HD 127617.

It is a pleasure to acknowledge that many of the Lick observations were made by Dr Jack E. Forbes, a graduate student at the time in the Department of Astronomy at the University of California, Berkeley.

Calibrated spectrograms at Hα of 566 stars in 29 young galactic clusters led to the detection of 41 Be stars in clusters. Using cluster membership we have inferred ages and intrinsic (B – V) colors of Be stars to permit a discussion of their evolutionary states.

Rotating stars can become Be stars in their early hydrogen burning evolution away from the main sequence. Both the fraction of stars showing hydrogen emission and the strength of emission appear to vary little during the first 80% of post main sequence evolution. However, at the onset of gravitational core contraction, both the fraction of stars showing emission and their mean emission strength undergo a fourfold increase. Many stars in the core contraction phase develop an intrinsic (B – V) excess of 0.15 mag. due to the H− free-bound continuum radiation. Because of the great strength of Hα emission and the short time duration of the effect, the extreme Be stars would be excellent probes for studies of spiral structure and would also serve as probes for studies of ages and distances of extragalactic systems.

Analysis of the corrected colors of the Be stars in clusters suggests that the Lucy and Solomon (1970) mechanism for reduction of effective surface gravity by ultraviolet resonance line scattering is probably important for the hotter Be stars. However, a discrepancy exists for the cooler stars between the predicted and observed colors.

At the Haute Provence Observatory, a survey of the Be stars brighter than the seventh magnitude was initiated by Herman and her collaborators twenty years ago.

o And is a shell star which has lost its shell characteristics several times since 1897. It exhibits short time scale light variations with a period of about one day. Its binary nature is doubtful according to Olsen (1972). We have detected rapid spectroscopic variations in October 1973 when o And had no envelope, that is, when it appeared as a ‘normal’ B star. Spectrograms of 7.2 Å mm−1 dispersion were obtained at the Haute Provence Observatory with exposure times ranging from 12 min to 52 min.

Photoelectric spectrum scans of Be stars were analysed to study the short-term variations of Hα emission strengths. In ζ Tau, α Col, PP Car, and δ Cen, definite variations of a few percent with time scales of 1 to 3 minutes were found. These variations do not exhibit any periodicity.

The paper describes the results of stellar photoelectric observations made with a small telescope and a 3 1/2 Å wide filter centred at Hα.

The method proved to be a sensitive one for detecting change in the spectrum but did not allow differentiation between changes in the Hα profile and wavelength changes due only to the Doppler effect. It was possible, however, to recognise and qualitatively to eliminate the Doppler changes due to the Earth's Annual Motion.

Of ten Be stars examined three, although in strong emission, showed little change over the periods of observation, the longest of which was 9 months. The other seven showed large and random changes.

Warnings are given about the difficulty of tuning such a filter exactly to Hα and the need to monitor accurately the filter for long term changes in passband and throughput.

The results of near infrared spectroscopic observations of over 120 Be stars are presented. In particular, the O I lines at λ 7774 and λ 8446 Å and the Ca II triplet (λ 8498 Å, λ 8542 Å, and λ 8662 Å) are discussed. The presence of emission at O I λ 7774 Å is shown to correlate with Fe II emission. Shell absorption at this line is found to be similar in appearance to that found in the photospheric spectra of supergiants. The O I λ 8446 Å line, the result of a fluorescence with Lyman β, is briefly discussed in connection with Be envelope dimensions. The paradox presented by the presence of strong Ca II triplet emission without H and K emission is discussed. The implications of this phenomenon on the structure of Be envelopes and possibly the binary nature of some Be stars is discussed.

Evidence indicating the presence of cool companions to the Be stars HD 218393 and HR 894 is presented.

Rectified intensity profiles of the Hα and Hβ Balmer lines of the shell star 4 Her were studied on 86 coudé spectrograms taken during 1969–1974. The central intensities of both lines vary periodically with the period of velocity changes equal to 46.194 days, reaching two maxima and minima each cycle. Radial-velocity curves of individual Balmer lines differ systematically one from another, the amplitude of the Hα variations being largest. The V/R ratio of the Hα emission peaks varies in phase with the velocity changes. The velocity of the Hα emission is found to be almost invariable. A model is suggested to explain the observed variations in which 4 Her is considered to be an interacting binary. The full paper will appear in Bull. Astron. Inst. Czech.27, No. 1 in 1976.

Radial velocities of high-dispersion spectrograms of ζ Tau, obtained mostly during the last observing season, indicate the following facts: (1) Several distinct velocity systems are present in the Balmer lines, not a continuous progression. (2) The main absorption component of the Hα line does not share the long-term velocity variations, its velocity being always rather close to the systemic velocity + 22 km s−1. (3) The pseudoperiod of the long-term velocity variations has changed and is probably decreasing.

In an attempt to shed some light on the origin of the material in the envelopes of Be stars, surface gravities were determined for 30 objects by comparing their observed profiles of Hγ and Hδ with those computed from the Princeton model atmospheres and the VCS theory of hydrogen line broadening. The program stars are predominately well-known Be stars and display a wide range of envelope spectra and v sin i. The mean and range in log g for the Be stars appear to be identical to that obtained from a similar analysis on non-Be stars. No correlation was found between log g and Hα emission strength or the strength and/or presence of emission of Fe II, O I λ 7774 Å, or the infrared Ca ii triplet. The suggestion made by Schild (1973) and Schild et al. (1974) that the extreme Be stars are in the post main sequence phase of rapid core contraction is weakened by the fact that there are several members of the class which have log g ≥ 3.8. All shell stars considered in the program appear to have low values of log g (≤3.5). Some possible explanations for this occurrence are discussed.

Two unstable shell stars, ζ Tau and 48 Lib, are compared. In the Herman-Rojas (1974) and Rountree-Lesh (1968) classifications, both stars have luminosity class III. These stars show similar features in the variation of the intensities, the profiles, and the radial velocities of the shell lines. An attempt has been made to explain the evolution of these unstable features.

A model for the shell star HD 50138 is inferred from the observation of the continuous spectrum from 0.14 to 10 μ.

HD 20012 (f1 Cyg) fait partie des étoiles des premiers types B. Nous l'avons classée O9 V?, Nina Morguleff l'a classée B1 IV et Janet Lesh B1, 5 V). Elle fait partie de notre programme d'observation qui concerne les étoiles Be de magnitude ≤7 et de déclinaison ≥ −15°. Parmi les étoiles les plus chaudes de ce programme, on peut signaler au moins deux catégories.

The appearance of singly ionized iron emission lines in the spectra of early type stars is studied, and the results of a spectroscopic investigation of EW Lac and other Be stars are given. We also discuss the atomic processes of excitation of Fe ii in the stellar envelopes using a two-parameter diagram W, NeT−1/2e.

The well-known shell star Pleione is also an interesting variable star, BU Tau. Its variability was first discovered by Calder (1937) from photoelectric observations in the 1930's. The light curve of this star from the end of the 19th century to 1954 was studied by Binnendijk (1949, 1955). Botsula and Sharov (1959) and Sharov (1961) added some old and new observations and continued the light curve up to 1959.