The accretion of gas on the surface of a neutron star is a source of energy, which can be converted into X-rays radiation. Accretion can ascertain a far greater lifetime of an X-ray star than the radiation of internal energy.

The source of gas could be a cloud of intergalactic gas. If the neutron star is a component of a binary system, the gas could come from the other (non-neutronic) star.

The idea of accretion and evaluation of energetic balance were given by Zel’dovič (1964) and Zel’dovič and Novikov (1964, 1965) and Salpeter (1964). Shortly the idea of accretion was mentioned by Šklovskij (1967).

The motion of gas and radiation spectra idealised case of spheric symmetry are studied now by N. Shakuro and Zel’dovič.

Evidence for a diffuse, isotropic non-terrestrial component of X-ray radiation has been accumulating since the very beginnings of X-ray astronomy. This evidence now seems secure beyond reasonable doubt, especially in the light of recent experiments, e.g. one in which a very clean separation from the cosmic ray and terrestrial background has been achieved (Pounds, 1967), and another in which a counting rate proportional to the solid angle of the sky subtended to three different detectors was found (Matsuoka et al., 1967).

In this very brief review I wish simply to discuss the current ideas on the mechanisms for the production of the background and the extent to which the present observations can discriminate between them and to consider the information obtainable from the very soft X-ray region, which so far has been relatively little exploited.

My task is to outline our present observational knowledge of the occurrence of lithium in stellar atmospheres. On account of the limited time, I shall not attempt to include a description of the situation in post-main sequence stars. Also for shortness of time, forgive me if I do not stop at each point to give due credit to the astronomers who have contributed to that topic.

The youngest stars we know which are cool enough to exhibit neutral Li are the T Tauri stars, which are in the early stages of contraction toward the main sequence. All these objects that have been adequately observed are very abundant in Li: the range is between 50 and 400 times the so-called ‘solar abundance’, a convenient unit which corresponds to a H/Li ratio by number of atoms of about 1011. It is significant that the average Li content of chondritic meteorites, which are often considered to be samples of the non-volatile, unprocessed material of the original solar nebula, is about 150 on this system, within the range observed in the TTauri stars.

Herbig’s extensive work on lithium abundances in solar-type stars on and near the main sequence, suggested to him that for these stars the lithium abundance might be some simple, decreasing function of age. If this were the full story we should expect a low lithium abundance in subgiants of about solar mass. However, already in Herbig’s work there were one or two stars with lithium abundance about 10 times the Greenstein-Richardson solar value whose trigonometrical parallaxes showed them to be possible subgiants. Other subgiants with high lithium abundances have been found by Wallerstein and at Pretoria. These results suggest that solar-mass stars are able to synthesise fresh lithium in or just prior to their subgiant phase.

Since a considerably higher temperature is necessary to consume Be atoms by (p, α) reactions than Li atoms, it has been thought that many clues to the understanding of the synthesis and destruction of the light elements could be found by studying Be abundances and the Li/Be-abundance ratios. Calculations have been made by Bodenheimer (1966) to determine the amount of convective depletion of Be during pre-main-sequence evolution. He found that a star reaching the main sequence at K5 would have destroyed 15% of its original Be; a K2 dwarf would have depleted its Be by 2%. For early K stars main-sequence depletion subsequent to this would be very slight. If there are no complicating effects such as magnetic fields and unusual patterns of circulation of stellar material, the observed Be abundances in stars of type G and earlier should be the initial pre-stellar or proto-star abundances. The BeII resonance lines lie far in the ultraviolet at λ 3131. But despite the very difficult nature of the observations, there are now Be abundances determined in about 60 dwarfs and sub-giants of types A, F, and G and in 2 K giants.

The determination of the lithium abundance in the solar atmosphere is essentially based on the LiI resonance doublet at λ 6707·761 and 6707·912 Å. These two lines form a very faint absorption feature, the central depth of the stronger component being of the order of 1% of the continuum. The violet component, which is also the stronger of the two, occurs near the red wing of a faint solar line of unknown origin, and the lines appear to be blended with other faint lines including possibly the doublet of the Li6 isotope (the isotopic shift being 0·160 Å). No other line of LiI has been detected in the Fraunhofer spectum of the undisturbed solar disk. This is nothing surprising, because practically all lithium is expected to be ionized in the photosphere on account of its low ionization potential (Xion = 5·37 e.v.). In sunspot spectra the lower temperature reduces the degree of ionization of lithium and causes a strengthening of the LiI lines. In fact, the LiI resonance lines which appear as a very faint absorption feature on disk spectra are about 50 times stronger in spot spectra. Furthermore, the very weak feature at λ 6103·6 Å was identified by Dubov (1964) and by Schmahl and Schröter (1965) as due to the 2s 2S–3d 2D transition of LiI. Both the resonance doublet and the faint feature at 6103·6 Å have been used by the above-mentioned authors to derive the lithium abundance in spots.

The difference between abundances of such elements as Li, Be, B in active regions on the Sun and in the undisturbed atmosphere, if it exists at all, could be considered a very important indication on the possible nuclear reaction in active regions and on the rate of diffusion of elements inside magnetic fields of sunspots. One of the authors from the considerations of λ6707 Å has an estimated (Dubov, 1955) Li abundance in the undisturbed solar atmosphere as log N(Li) = 0·93 (taking log N(H) = 12·0), which is close to the later results (Goldberg et al., 1960). In sunspots we can expect that the first doublet λ6103 of Li diffuse series appears. The preliminary consideration (Dubov, 1964) showed that Li abundance in sunspots could be several times larger than the above estimate for the undisturbed atmosphere. This result forced us to consider as carefully as possible the region of λ6103 in eight spectra of four sunspots taken with the aid of the echelle-grating spectrograph of the Crimean observatory with a dispersion of 1 Å/mm and measured resolving power 0·03 Å. The consideration of the Zeeman pattern of this line shows that at the field strengths we had in sunspots (2000–3000 gs) the line should be diffuse band with width ~0·23 Å, and taking into account the broadening of the line the halfwidth of the line has not to be smaller than 0·25 Å.

In 1955 Fowler et al. (1955) suggested that the elements Li, Be and B could be produced at the surface of some stars by the spallation of CNO induced by electromagnetically accelerated particles.

Since that time many observational data were gathered, not only on the lithium elemental abundance (Hunger, 1957; Bonsack and Greenstein, 1960), but also concerning the isotopic ratio 7Li/6Li (Herbig, 1964; Wallerstein, 1965). Then, the abundance of beryllium in many stars was measured (Bonsack, 1960; Conti and Danziger, 1966; Merchant, 1966) and the problem was, and still is to explain both 7Li/6Li and Li/Be ratios. In addition to spallation, different possibilities may be invoked, such as the inheritance from galactic gas enriched with matter remaining from a previous generation of stars, or from the big bang; but which can be checked most easily at the present time is the spallation assumption. Indeed, simultaneously with the development of the observations on stars, a program for measuring and calculating the production rates of the light elements has been in progress at Orsay. In 1960 the first experimental result concerning the production of 6Li in 16O by 156 MeV protons was obtained. But it was only in 1965 that the number of experimental and theoretical data on spallation was sufficient to make a valid comparison with abundances observed on the stars as well as in the solar system or in cosmic rays.

I want to state a few questions which are raised by the observations of stellar Li and Be, and to see how far we can go in giving answers.

I do not believe that we can, at the present time, give very definite answers to most of these questions and build up a unique model in which all observations are properly understood. We are still faced with different possibilities which can not be excluded, but can only be classified according to their degree of ‘plausibility’. Although we are led to choose one of these possibilities in the light of present data, it seems wise to keep the other possibilities in the back of our mind, and to assess their merit. Considering this, a few alternate models will be described and discussed.

The observational status of the problem of lithium abundances has been thoroughly detailed in the preceding papers in this symposium, and it is clear why we must consider how matter is mixed from outer convection zones to inner, hotter regions. The need for appropriate mixing mechanisms has also been nicely brought out by Herbig and Wolff, and Böhm has summarized the role of convective mixing. Conventional mixing-length models for the outer convection zones seem to give qualitatively reasonable results for the depletion during pre-main-sequence contraction but do not completely account for the observations, and it seems inescapable that main-sequence depletion of lithium must be considered. I shall therefore simplify the discussion of mixing by concentrating on main-sequence models in the following outline of some possible mixing processes, though most of the remarks to be made should apply generally to other phases. I shall also pretend that there is one principle mechanism (or combination of them) that must be found, though stars in different evolutionary phases, or with different masses, may deplete lithium quite differently. Further, I shall use the Sun as an illustration in general since we know some important details about it that are not always known for other stars.

This report consists of a brief summary of the status of a number of programs of an international nature which are now in progress at various observatories of the world. These programs have received strong endorsements from Commission 8 of this Union, as well as from a number of international organizations interested in accurate positions and motions of the stars to the 9th magnitude of brightness. Among the organizations and conferences which have explicitly endorsed the programs are:

(1) The Symposium on Coordination of Galactic Research, Groningen, 1953.

(2) Conference on the Problems of Astrometry, Evanston, 1953.

(3) Astrometric Conference, Pulkovo, 1954.

(4) Conference on Astrometry, Brussels, 1955.

(5) International Union of Geodesy and Geophysics, 1954 and 1963.

(6) International Astronomical Union, 1955.

(7) Second Astrometric Conference, Cincinnati, 1959.

(8) First Inter-American Conference on Astronomy, La Plata and Cordoba, 1959.

(9) Meeting on Problems of Astrometry and Celestial Mechanics, La Plata, 1960.

(10) Committee on Space Research, COSPAR, 1966.

Absolute determinations of star positions using transit and vertical circles according to the methods developed at Pulkovo are being made at Pulkovo, at Nikolaiev, at the Golossejevo Observatory of the Ukraine Academy of Sciences, and at Santiago de Chile. At this latter observatory Soviet and Chilean astronomers have completed a series of observations of both bright and faint fundamental stars with a photographic vertical circle, and these observations are now being reduced at Pulkovo. Absolute determinations of right ascension will be made at the Cerro Calan Observatory using a large transit instrument recently constructed at Pulkovo. Further details of some of the Southern observations are given in the following paper to be presented by Dr. Anguita.

Observations of the Southern Reference Star (SRS) and Bright Star (BS) programmes were started in January 1963 with the Repsold meridian circle of the Santiago Observatory. Up to July 1, 1967, a total of 86725 star observations arranged in 1520 series had been observed.

In order to investigate the instrumental system, series of fundamental stars between +41° and −69° SP were included in the observing programme. A total of 214 such series were observed, 123 being with ‘clamp East’ and 91 with ‘clamp West’. The number of individual star observations in these 214 series was 7344. These observations were reduced using Bessel’s formula

α = T + (u + m) + n tan δ + c′ sec δ,

where u denotes the clock correction and c′ includes the collimation error, diurnal aberration and the contact width.

Absolute proper motions have been determined for 14000 stars in the 82 areas indicated in Figure 1. These proper motions result from measures made on 105 plate pairs taken with the Pulkovo normal astrograph which has a focal length of 3·5 m. The plates, which had an average epoch difference of 22·4 years, were measured film-to-film in a Repsold measuring machine. The number of galaxies used for defining zero proper motion was 271, and among the stars measured were 1283 from the AGK3. The probable error of a determination of proper motion from one plate pair was

The images of the galaxies were estimated as having an average quality of 6 on a scale of 10. (Deutsch et al., 1955).

A fundamental catalogue compiled from independent and differential observations of stellar positions at various epochs provides the fundamental system of positions and proper motions of the stars. The system ought to be compiled in such a way that it represents the inertial frame of reference as accurately as practicable. Its direct purpose is to serve as the reference system of positional astronomy. It should fulfil the requirements of astronomical research, in particular in celestial mechanics and galactic research, as well as the demands of the astronomical determinations of time and of latitudes. It has also to fulfil the requirements of geodesy and navigation. For all these purposes the system must be uniform over the whole sky (free from regional systematic errors), and it has to be uniform over a large range of stellar magnitudes (free from errors depending on the magnitude of the stars).

Previous determinations of precession on the basis of proper motions resulted in corrections to Newcomb’s value of lunisolar precession in the interval per century. From a compilation of several determinations and their reduction to the average of the systems of FK3 and N30 Morgan and Oort (1951) obtained:

per century, where A and B are Oort’s constants. Values in the system of FK4 were derived by Bošniakovich (1965), who applied differential corrections to previous results on the basis of the systematic differences FK4–FK3.

In many problems of galactic kinematics we need to know the proper motions of faint stars, as well as bright stars, within some well-defined reference frame. Generally speaking, the accuracy required is greater for faint stars on account of their larger distances. Techniques at present used for determining proper motions vary according to apparent magnitude, and it is clearly desirable that all methods should give results which are capable of reduction to a common frame.

Problems in galactic research for the solution of which highly accurate absolute proper motions are required are manyfold. The following selection seems of particular interest at the present moment. Problems for which only accurate relative proper motions are required are not considered here.