It was the feeling of the Scientific Organizing Committee that 50 years after the introduction of the concept of stellar populations it might be desirable to attempt to rationalize their classification and terminology. Toward this end a panel discussion was organized, which was followed by a period of audience participation. Although not all those who contributed to the latter section turned in a written version of their remarks, that section is reported as well as possible here.

Of all the subfields which make up the discipline of astronomy today stellar populations must surely be one of the most difficult to explain to the lay person. From the first day of this meeting I took the notion that this was the invention of a charismatic individual with a particular genius for bridging the gap between stellar and extragalactic research, a bridge which is probably still very relevant to maintain today.

The elemental abundances of late B and early A stars should be among the most reliably derived as these stars have well known atmospheric physics, and non-crowded optical spectra with lines of many species with good oscillator strengths. We obtained elemental abundances of sharp-lined stars from high dispersion, S/N > 50, spectra taken at DAO and KPNO. We performed fine analyses using ATLAS6 and more recently ATLAS9 model atmospheres. Our papers (mostly in MNRAS) contain consistent studies of about 35 B3 to F7 superficially normal, Mercury-Manganese (HgMn), Am, and Field Horizontal Branch (FHB) A stars. The abundances of the normal main sequence B, A, and F stars are indicators of Galactic metallicity during the most recent epoch of Galactic history.

Formation of the flare stars database (FSDB) continues (Tsvetkov et al. 1993). Since the beginning of systematic monitoring up to the present 1608 flare stars have been discovered with more than 3107 flare events observed for more than 9313 hours. (Repeated flares for stars from the solar neighbourhood, as well as the time used for their monitoring, are not yet considered).

A new set of spectral indices computed for the Mg feature at 5175 Å is presented. The spectral indices were computed using the most recent grid of model atmospheres and programs to produce synthetic spectra (Kurucz 1993; CD-Rom 18) for the temperature range 4000-6000 K, and gravity interval log g = 1.5-4.5. The indices reported here refer to solar chemical abundance.

Some years ago a complex program of studying the main meridional section of the Galaxy was initiated with the aim of improving our knowledge of spatial and kinematic characteristics of stellar populations. To classify stars, objective prism stellar spectra (D = 166 A/mm at Hγ), are used. The field diameter is 4° 50′, the limiting photographic stellar magnitude is about 12m. Our automated quantitative spectral classification of F-K stars applies criteria evaluation and is based mainly on the SDR package for spectrophotometric data reduction (Malyuto, Pelt, Shvelidze, 1993) and the CTATEC-2 package for the definition of a multiple linear regression model “criteria values versus main physical parameters” (Malyuto, Shvelidze, 1989). Our regression model was based on the final sample of calibration stars containing 95 standard (bright) stars and 96 program faint (8m < B < 11m.6) stars from our areas near the North Galactic Pole. The standard deviations of our calibration with the use of all data taken together are ±0.015 for log Teff, ±0m.96 for Mv and ±0.25 for [Fe/H]. These results are encouraging for application of our method to a large set of Abastumani objective prism spectra.

We present the first models of massive stars, where the interior structure is coupled to a spherically extended, metal line blanketed non–LTE atmosphere with a wind. This is accomplished with the Geneva stellar evolution code and the non–LTE atmosphere code of de Koter et al. (1993). The method described in Schaerer & Schmutz (1994) is used to treat line blanketing. A first account of this work is published in Schaerer et al. (1994).

Pulsating A-F variables include all the stellar types listed in teh Table, as well as the pulsating white dwarfs. Stars near the zero-age-main-sequence have faster rotation velocity, which slows as expected with age (Villata 1992) and a smaller amplitude of light variation, so we suggest that rotation velocity be considered in Population classifications. Also, in the Galaxy, the galactic rotation constant A is related to stellar age T by: A(kms−1kpc−1) = (−2.4±0.8)T(109yr) + (32±2) (Kharchenko 1992). The linear rotation velocity is also a function of the Z coordinate of the object inside the Galaxy: the mean Z-gradient is −10kms−1kpc−1 (Malakhova & Petrovskaya, 1992). Thus the population is strongly correlated with the rotation velocity and the evolutionary age.

The previous calibration of the Strömgren photometric indices in terms of spectroscopic [Fe/H] values was performed by Schuster and Nissen (1989) with the application to their sample of calibration stars a mathematical package handling the regression of [Fe/H] values against the (b – y), m1, and c1 indices, and their combinations. To provide a similar regression we applied a quite another independent statistical method (notably that of of stepwise linear regression) to the same sample and also to other available calibration samples of F stars. We conclude that in our case the quality of calibration was slightly better and the analytical form of calibration equation was much simpler.

A photometric reddening-free calibration for [Fe/H] valid for giant and supergiant stars of intermediate temperature, has been obtained using the Strömgren uvbyß system. Galactic supergiants, supergiants in Magellanic Clouds and galactic metal deficient red giants with spectroscopic determinations of [Fe/H] were used as calibrators. The calibration can be used to predict [Fe/H] with an accuracy of 0.33 dex and shows the potential of supergiant stars as tracers of iron abundances in other galaxies.

We are automating the multi-parameter classification of stellar spectra within the MK system. The increased rate at which spectra are gathered by increasingly larger telescopes, fibre systems and efficient CCD detectors necessitates the need for automation of stellar classification. As well as greatly speeding up this process, an automated scheme would give reproducible results, would arguably be more objective, and would give quick and useful reporting of classification uncertainties.

We present the results of a general programme aimed to study the effects of metallicity on the temperature scale of late main sequence stars (F0-K5). A sample of approximately 400 stars with published UBV(RI) and ubvy – β photometry has been collected from the literature. A three years campaign of observations (Alonso et al. 1994b) was carried out to obtain JHK photometry in order to apply the InfraRed Flux Method (IRFM) to derive effective temperatures. The effect of metallicity on Colour-Colour IR diagrams is discussed. The absolute flux calibration in the IR was revised in Alonso et al. (1994a). The effect of metallicity on the bolometric correction has been studied in Alonso et al. (1995) in order to derive bolometric fluxes. The temperatures have been derived by applying the IRFM using new Kurucz models. Teff = f(Colours, [Fe/H], log(g)) relations are obtained for dwarfs covering the ranges 4000K ≤ Teff ≤ 8000K, 3.5 ≤ logg ≤ 5.00, +0.5 ≤ [Fe/H] ≤ −3.00 which expands considerably the database of previous works. These relations are used to check atmosphere models through the analysis of UBV RIJH Kubvy – β synthetic photometry in combination with the IRFM. The transformation from the theoretical HR diagram into an observational one is analyzed with the new relations. The influence of these points on the study of the evolution of the Galaxy is briefly discussed.

There are a number of stars with very high pecular velocities in the Galactic halo within the region of about 1.5 kpc from the Galactic plane. At least some of these stars could be the result of an escape process in multiple stellar systems located in the Galactic disc.

We computed the dynamical evolution of hierarchical triple stars in which both orbits are initially circular, and determine the lower limit to the ratio of periods (outer/inner) for which there is dynamical stability. We found for some mass ratios resonance-like behaviour that occurs in a limited range of initial period ratio. Some resonances are ‘disruptive’; that is, for a small range of initial period ratio we find that the system is not able to settle down to a quasi-steady hierarchical state, but instead disrupts. However, below as well as above this disruptive range there are considerable ranges of initial period ratio where the hierarchical state appears to be stable, at least for the length of integration time we took which was sometimes as much as 10,000 outer orbits. The mass ratios are identified for which different types of unstable behaviour, such as an escape of the distant body without exchange, many exchanges in the limited space without escape, formation of new long-live hierarchy, or an escape of one body after a few exchanges, occur for ratios of periods slightly below the limit of stability. We discuss the relevance of the above behaviour to observed close triples, the closest of which is λ Tau (period ratio 8.3).

We study the dynamics of extended shells of relatively low-mass particles around and inside the orbit of two heavy centres of gravity (a binary) by computer simulations. The binary components are surrounded by N=16000 small mass particles in a uniform random distribution on a few spherical envelopes with different radii expanding with some velocity. Some shells are inside the orbit of the binary.

Stars show a marked tendency to form the systems of different multiplicity starting from the smallest systems as binary and triple stars up to clusters with a significantly larger number of objects (N ~ 107 for globular cluster s). Different investigators have used different methods of binary stars identificati on but modern observations give a frequency of binary and multiple stars in the Galactic field up to 70% Binary and multiple stars are also often present within star clusters. It is therefore very important to be able to identify such systems as rather isolated substructures in which the dynamical evolution is not significantly affected by other stars (at least, during some considerab le time interval).

We discuss the determination of the expansion of the Orion association. Previous estimates of expansion in associations have employed simplified linear approximations which only required the observation of proper motions (cf. Lesh, J.R., ApJ, 152, 905, 1968). We have used proper motions obtained by Smart (PhD Thesis, Univ. of Florida, 1993) with previously obtained membership criteria to investigate this hypothesised expansion without conclusive results.

This study is based on the catalogue of 69 open clusters by Barkhatova et al. (1987) and cluster distances taken from Hagen (1970). Most of the dimensionless characteristics discussed here are insensitive to the distance scale. Galactocentric cylindrical coordinates of clusters and the corresponding velocities have been calculated with the accepted galactocentric distance of the Sun R⊙ = 8.2 kpc and the galactocentric velocity of the Sun sR = −9 km/s, sθ = 228 km/s, sz = 7 km/s.

Johnson UBVRI CCD-photometry was performed for 18 fields (3′×4.4′) in the association Cepheus OB3 using the 1.23m telescope of CAHA at Calar Alto (Almería, Spain). A total of 1055 stars was detected and the sample is complete down to V ≈ 19mag, although only 130 stars were measured with the five filters.