A series of plates is being taken with the U.K. 48-inch Schmidt Telescope at Siding Spring on a field centred on the star GC 1110 near the South Galactic Pole, in order to detect any stars brighter than about mpg = 18, which may have significant trigonometric parallaxes.

A search list of more than 16, 300 objects in an area 4¼° × 4¼° has been formed from a preliminary scan on the Galaxy machine at R.G.O. Prom 1974 January to 1976 January some 30 plates have been obtained, nearly half of which have been measured in each of two orientations on Galaxy; each measuring run takes about 16 hours.

The Observatory of Strasbourg will participate in the future in the general radial velocity survey which is in hand at the Observatories of Haute-Provence and of Marseille with FEHRENBACH’s Objective-prism astrograph. A programme of this type, providing a large number of data, is particularly suitable at the place where the “Data Center” is growing up.

A radial velocity survey is a vast programme which needs many years to be fulfilled. It is natural that one starts with the most interesting areas in the sky, one of them being the area near NGP. In 1976 twenty fields were covered, each by one plate, in that direction. The whole programme covers 70 overlapping fields in an area of 100 square degrees centered at NGP. The limiting magnitude shall be almost the 12th.

This is a progress report of a project designed to find the component, perpendicular to the Galactic plane, of the gravitational potential of the Galaxy. The principle is to measure the radial velocities and distances of a large number of K-giant stars near the North Galactic Pole. My student G. A. Radford is masterminding the project; collaborating with us are Drs. J. E. Gunn of the Hale Observatories and L. Hansen and K. Gyldenkerne of Copenhagen.

We have measured the radial velocities of all the HD stars of type KO and later, and many of the G5 stars, within 15°of the Galactic Pole, using the Cambridge photoelectric spectrometer. In addition, we have observed all the stars classified as K giants by Upgren in his declination zones 25° to 31°, using the spectrometer on the Hale telescope. There are about 900 stars observed altogether, including about 200 Upgren stars, running down to twelfth magnitude or so, which are not in the Henry Draper Catalogue. To determine the distances of all these stars we are now trying to determine the absolute magnitudes by narrow-band photoelectric photometry in the Copenhagen system. Most of the observations have been made, thanks largely to the very generous grants of observing time given by the Hale Observatories earlier this year; but the reductions have only been completed for about 300 stars (including 244 K giants) which were observed last year at Kitt Peak, and the present, very preliminary, discussion is based on those stars alone.

Several Lick 3-m prime-focus plates of Selected Areas 57, 68, and 51 (taken by I. King) in B, V, and R have been measured for stellar magnitudes down to the plate limit by L. Hinrichs and King, and are currently being measured by Chiu for proper motions (several hundred stars per plate) with the Berkeley PDS microdensitometer. Prime-focus plates are also available from the Hale 5-m and Mayall 4-m telescopes, giving an overall baseline of 20 years. Work so far indicates that on the Lick plates stars brighter then V = 19 can be measured to within one micron standard error; the error becomes unacceptably large for stars fainter than V = 20. A large number of stars bluer than B-V =0.4 show proper motion and are therefore excellent candidates for white dwarfs. For 0.4 ≤ B-V ≤ 0.8, the proper motion stars are expected to be predominantly subdwarfs.

The frequency distributions of the stars in V and B-V for the three fields are being analysed by Brooks; the fields are advantageously placed for study of the density distributions in both the disk and the halo. These data should allow the halo stars to be studied out to a distance of 10 to 15 kpc, as well as a determination of the degree of flattening of the halo. Also, a study will be made of the z and ῶ density gradients in the disk, and the luminosity function of disk stars.

Swedish investigations on stellar space distribution generally are based on the Uppsala - Stockholm spectral classification system. In this system the properties of the stars are derived from objective prism spectra by means of narrow-band photometry. The main criteria are the hydrogen-line intensities, the intensity of the K line, the intensity of the G band, the break at the G band, and the intensity of the blue cyanogen band (for a description of the system see e.g. Ljunggren and Oja, 1961), when possible supplemented by a measure of the Balmer discontinuity (Rydström, 1976). The method yields absolute magnitudes with a dispersion of about and intrinsic colours (B-V)° with a mean error below for most kinds of stars.

Three investigations are relevant to the subject of this Joint Discussion.

A short account was given of results recently published (Hill et al. 1976; Hilditch et al. 1976a, b) of a spectroscopic and photometric study of 310 A0-F8 stars within 15° of the north galactic pole. Using these data, it has been found that δmo for the programme stars is constant out to 250 pc and corresponds, via Crawford’s (1975) calibration, to the solar value of [Fe/H]. The distribution of radial velocities of A0-F8 stars is asymmetric and may be interpreted as showing a broad distribution of mean −7 kms−1 and dispersion σ ~ 11 kms−1 together with a superposed narrow distribution of mean 0 kms-1 and dispersion σ ~ 3 kms-1. The narrow component originates from a group of 37 stars which appear to be kinematically and spatially associated with the Coma cluster. Only 14 of these 37 stars are recognised members of the Coma cluster. Removal of these 37 stars from the sample results in a distribution which matches closely the W component distribution of A0-F8 stars in Gliese’s (1969) catalogue. Thus Oort’s “well-mixed” hypothesis seems to be confirmed out to 250 pc. Much additional work on the space motions is required before the hypothesis of the ‘Coma group’ can be thoroughly tested. Results of this analysis and a preliminary value for the total mass density in the solar neighbourhood is expected to be completed during the next few months.

Since the time allowed per paper at this joint discussion is short only one aspect of the stellar distribution at high galactic latitudes will be presented here, namely the distribution of field horizontal-branch stars (FHB) in the galactic halo. First, the method by which FHB stars are found will be described. Second, the density distribution of FHB stars will be compared with that of the RR Lyrae stars.

As part of a general program to study the stellar density distribution perpendicular to the galactic plane a number of possible FHB stars has been discovered. In each survey area objective prism plates (at a dispersion of 280 Å/mm. to a limiting magnitude of V = 14) are taken with the Michigan Curtis Schmidt telescope at Cerro Tololo Inter-American Observatory or the Schmidt telescope at the Warner and Swasey Observatory. A set of direct plates are taken also to obtain photographic magnitudes for the stars with spectral classifications.

Recently Dessureau and Upgren (1975) redetermined the velocity distribution of giant stars in the north galactic pole direction using Upgren’s (1962) catalogue and Oort’s (1960) determination of K(z). The velocities were assumed to be represented by n Gaussian distributions with no further constraints imposed. The velocities are well represented by three such distributions whose properties disagree with those found by Oort. A larger number did not improve the stability of the solution. Without radial velocities, however, they could not redetermine the K(z) force itself.

Stock and Osborn (1972, 1973) have shown that objective-prism spectra taken with a conventional prism may be measured to produce radial velocities of sufficient accuracy for statistical purposes. They are determined by means of a third-order power series in both x and y coordinates. The same spectral-line measures can also yield positions such that proper motions can be determined if first-epoch positions are also available. For many stars, both tangential and radial velocities can be obtained with about the same error which is of the order of ± 20 km/sec. The field distortions caused by the prism are large but are constant and predictable to the degree that measured residuals are similar in size to those for direct images (Stock and Upgren 1968). A survey of a high-latitude zone between −30° and −35° in declination is underway and a catalogue of about 3000 stars has already been compiled by Stock. For each star, the catalogue lists an accurate 1950 position, spectral and luminosity type, apparent photographic magnitude, relative radial velocity and its weight, and the number of plates on which the star was measured.

A matrix method involving eigenvector expansion is used to solve the “fundamental equation” of stellar statistics. This method is applied to M dwarfs and K giants.

The motion of G8-K5 III, IV stars near galactic poles (|b|>60°) was investigated. The main attention was paid to motion in the z-direction. The spectral types, visual magnitudes, colors and radial velocities (vr) were taken mainly from Schild (1973), Heard (1956), Blanco et al (1968) and Wilson (1953). To reduce the values vr by Heard to those by Wilson, the correction of -1 km/sec was made. The stars with variable vr and spectroscopic binaries were excluded from the investigation.

The papers presented this morning form a good illustration of the old adage that you should never publish a paper that is completely correct, for then you can write only one paper. But if you publish a paper with a lot of errors, you are sure to be criticized, and you can write a second paper to answer your critics, and if the heat proves to be too much you can write a third paper withdrawing everything.

First a bit of history. It is difficult to pin-point when exactly this question of the star density in high latitudes first came up, but I feel sure that it is Implied, to say the least, in the early work on faint blue stars by Malmquist (1927, 1938) and by Humason and Zwicky (1947). This morning Ivan King showed us a table of the changes in the frequency of occurrence among the stars in high galactic latitude as one goes to fainter and fainter magnitudes. This is exactly what I did in 1960 when I published a color-magnitude diagram for 4000 stars down to 19th photographic magnitude near the South Galactic Pole (Luyten 1960). These data had been obtained with the Palomar 48-inch telescope using Haro’s three-image method. I also made up a similar diagram calculated from what I thought were then the best available data on the luminosity and density functions. The conclusion of my analysis was that there seemed to be rather fewer M stars than had been expected, and a great many more stars with the color of F and G stars than expected. This idea was not popular at the time; hence this paper has been carefully ignored until these ideas were used by others twelve years later without reference to the 1960 paper.

In research in the galactic polar caps we may roughly distinguish two categories:

I. Research on problems concerning the galactic population at large distances from the plane, including the density and velocity distribution and the field of force at large z.

II. Research on problems concerning the population in the immediate neighbourhood of the sun; at high galactic latitudes it is easier to discriminate the nearest stars than at low latitudes - see, for in stance the investigations of the M dwarf population.

Most of the papers presented in part II of this Joint Discussion deal with the first category, and of these, the distances reached are mostly within a few Kpc. Classical investigations, some of them dating from more than thirty years ago, have provided our current knowledge of the force K(z) perpendicular to the galactic plane up to about 1500 pc and the large scale features of the density distribution. What emerges from recent and current work, is attempts to improve the knowledge of K(z) and to extend it to greater distances; and refinement of the density and velocity investigations by more and more detailed discrimination of population types. The dominating problem is that of the separation of the stars according to two basic parameters: age and chemical composition. Kinematic and space distributional properties for subgroups according to these two parameters will be the necessary elements on which a theory of the (local) evolution of the Galaxy is to be built. The vague notion, that metal abundance may be taken to be an equivalent of age is abandoned and replaced by the recognition that at any epoch, star formation may have occurred with a considerable spread in the resulting stellar metal abundances.

In the days following our Joint Discussion the Co-editors were joined by two of the participants, Dr. Ivan King and Dr. Uli Steinlin, in an effort to evaluate results reported as well as the problems posed for future work. As a result of these discussions at Grenoble the present listing was made. This list was presented to members of the Organizing Committee but since it was impossible in the last days at Grenoble to convene this group or to speak collegially with all participants it must remain the responsibility of the above-named authors. It cannot be a complete resumé nor can it presume to represent adequately the varied opinions expressed at Grenoble on 25 August. We hope here only to attempt a synthesis of certain evident results and to outline certain prospects for exploring further the exciting problems of structure and evolution in the galactic polar caps.

Welcome to this Joint Discussion on stellar atmospheres and interiors. For stellar structure, the atmospheres are of importance for two reasons, first as a source of information on effective temperature, gravity (or mass: luminosity ratio), chemical composition and dynamical effects, and secondly as an outer boundary condition for stellar models.

In the last few years there have been substantial advances in stellar atmospheres as a result of both improved theories and new measurements. In the particular field of abundance determination, non-LTE theory has at last come of age and provided a certain number of corrections that are both realistic and significant, for example in reconciling the neon abundance in B stars with that in nebulae and solar flare particles. Solar abundances have been improved by the provision of better oscillator strengths and better treatment of line broadening, generally with the effect of leading to still closer agreement with carbonaceous meteorites. Nevertheless, he would be a bold man who claimed to know the initial solar abundance parameters Y and Z to better than, say 25 per cent, even if we grant that the photosphere is a true sample of the initial interior abundances - an assumption that is now being questioned (again) in view of the solar neutrino problem.

As a star burns its nuclear fuel, its radius R and its luminosity L are modified. Its mass may as well be affected if the mass loss rate has a time scale comparable to the nuclear time scale; this is likely to occur for stars of very high luminosity. Currently, the change in radius R and luminosity L of an evolving star is described in the socalled theoretical Herzsprung-Russel diagramme with in abscissa the logarithm of the effective temperature defined by:

It is reasonable to suppose that the state of the convective zone underlying the atmosphere of cool stars correlates with the atmospheric kinematics. Following Iben (1967), as a star evolves from the red giant tip to the helium burning core phase, the convection rapidly decreases. Then the microturbulence can be expected to decrease too.

Abundance peculiarities in successive stages of stellar evolution are reviewed. Main-sequence stars show anomalies in lithium and, on the upper main sequence, the Am, Ap and Bp effects, which may be largely due to separation processes, and helium and CNO anomalies to which nuclear evolution and mixing could have contributed. Red giants of both stellar Populations commonly show more or less extreme variations among the C, N, 0 isotopes, sometimes accompanied by s-process enhancement, due to mixing out in various evolutionary stages. Detailed anomalies expected from galactic evolution are also briefly considered. Novae show strong effects in C, N, 0 and synthesis of heavier elements is displayed by the supernova remnant Cassiopeia A.