The advantages of making astrometric observations from space are reviewed, and a project for measuring angular separations between stars along great circle arcs of the order of 90°, by means of a telescope mounted on a TD 1 satellite is described. Positions with accuracy of the order of ±0″.01, parallaxes to ±0″.007 and proper motions to ±0″.005 yr-1 for some 40000 stars brighter than 9m.5 would be obtainable in a two year programme.

The Large Space Telescope (LST), planned to be launched in 1980, will give an opportunity for astrometric observations of high quality to be secured.

A detailed survey is presented of the potentialities of the LST for studies of proper motions, parallaxes, star clusters, double stars, perturbed motion stars, occultations, fundamental astrometry and solar system observations.

The proposed characteristics and design capabilities of the LST are tabulated.

The Automatic Measuring System (Lick Obs. Bull., No. 598, 1971) was built according to a design developed in 1959. Technological advances since then, particularly in small computers and in electronic component reliability, offered an opportunity for a substantial upgrading of the system. Most of the electronic circuitry has been replaced using modern components, and a PDP-8 computer has been added to the system. Punched card equipment has been replaced by magnetic tapes.

The new 48-in. Schmidt telescope of the U.K. Science Research Council is being commissioned at Siding Spring in Australia. Its potential both for extra-galactic and galactic research is enormous but this potential can only be fully exploited with an automatic measuring machine.

A new ‘Starscan’ measuring machine has been built by Optronics International, Chelmsford, Mass., to specifications drawn up by the U.S. Naval Observatory. The accuracy and speed of operation of Starscan have been optimized for the measurement of astrometric plates. The following features in the design and operation of Starscan are discussed: (a) the granite stages, (b) interferometric stage position encoders, (c) the star centroid and integrated opacity detector, (d) the image profile display system, and (e) the function of the computer in machine control and data manipulation. The performance of Starscan, in terms of operating speed and accuracy, is reported for astrometric plates.

Everybody who deals with the numerical determination of parameters from observations is familiar with the phenomenon that the ‘external’ errors of these parameters are invariably larger than their ‘internal’ ones. This phenomenon is traditionally explained by the presence of systematic errors.

Since it is virtually impossible to use a completely rigorous reduction model in any astrometric problem, an ‘interpolation’ model must normally be employed in which there are fewer parameters (unknowns) than in the rigorous one. It is shown that under these circumstances, the rms errors of the adjustment parameters are always underestimated, which creates the illusion that the results computed with these parameters are more accurate than they actually are.

A modification of the normal regression models used for the determination of star positions is described. Some preliminary results relating to accuracy are given.

The evaluation of stellar positions from overlapping plates by rigorous block-adjustment methods leads to a very large system of normal equations containing both stellar positions and plate constants as unknowns. These general normal equations can be reduced to a subsystem with banded-bordered coefficient matrix, containing only the plate constants as unknowns. In the case of a large photographic net covering a hemisphere or sphere with multiple overlap of the plates, the actual number of unknown plate constants may exceed 20000, depending on the adopted reduction model.

La réduction des clichés astrographiques nécessite l'emploi de formules dont le choix relève du problème de la représentation mathématique de phénomènes expérimentaux.

A photoelectric parallax scanner designed and constructed at the Yerkes Observatory to measure parallaxes and proper motions at the telescope is described. The concept of the parallax scanner is similar to Griffin's radial velocity spectrometer except that the parallax star and the field stars are monitored separately, thus making the technique purely differential. Observations with the 40-in. refractor under less than ideal conditions gave the error of a single measurement as ±4.5 μ (s.e.). Observations under clear skies and with improved drive and guidance should decrease the error by a factor of 2 to 3.

Two recent papers (Hershey; Gatewood and Eichhorn; Astron. J., 1973) pointed out that:

(1) A proper motion irregularity interpreted as orbital motion was found duplicated in another field.

(2) The discontinuity found at one telescope failed to be confirmed by others, and might be ascribed to mere instrumental errors.

After the duplication was discovered two years ago, further measurements have revealed an identical pattern in at least seven parallax stars. Yet the cause of the effect, and the reason for its absence in other cases, have not been located. Colour effects and filter problems appear to be ruled out. Indications for discontinuities occurring at other epochs were found. Thus, the question cannot yet be safely answered whether the small variations (one or two micron) of proper motions reported in some cases are spurious; in any case, great caution should be exercised in interpreting one-micron effects unless confirmation from another instrument is obtained.

For a precise reduction to the apparent places of the stars in a uniform system during the 19th and 20th centuries, the ‘Solar Coordinates 1800–2000’ by Herget (Astron. Papers14, 1953) may conveniently be used, because no coordinates of the Sun, referred to the mean equinox of 1950.0, are given in the Astronomical Ephemeris before 1930.

A maximum difference of 0″.0003 was found between the aberrations calculated from both the Astronomical Ephemeris and Herget's Tables for the period 1960–1969, taking into consideration the effect of the outer planets, which amounted to 0″.0109.

The effect of the inner planets on the aberration is estimated to be of the order of 0″.0001 at the most and the correction for the lunar term due to the change in astronomical constants is 0″.00002. It is recommended that the solar coordinates be calculated directly from Newcomb's formulae taking the effects of all the planets into consideration, but the effect concerned with the Moon can be neglected.

This has been a good Symposium. There obviously is a need to review now the problems of the accepted fundamental reference system of star positions and proper motions. The basic Fourth Fundamental Catalog (FK4) has been the reference catalog for the past 10 years. It needs updating and especially it should be made more directly usable for discussions of positions and motions referred to faint galaxies. In the preparation of the next catalog we should make use of radio galaxies as basic reference points for fixing precision stellar positions. The Symposium came at the right time! Radio Astrometry has burst upon the scene and it is essential that the optical and radio astrometrists should get to know each other and exchange views about the manner in which together we may work towards the establishment of a fundamental system of positions and proper motions more reliable than we have had in the past. There has also been much activity in the area of measuring proper motions of faint stars relative to galaxies and we obviously have to consider the best manner in which these valuable new contributions can be applied most usefully to basic astrometry. Our Symposium was held in the right place! It is high time that we should draw attention to the accomplishments and the future needs of basic southern hemisphere astrometry; Perth, Western Australia, is obviously a good place to discuss such matters. Finally we should discuss questions relating to instrumentation. Over the past decade there has been what could be called a ‘breakthrough’ in instrumentation, not only in the radio area, but also in the more traditional optical area of measurement by transit circles. There have been many new developments in the field of automatic measurement of photographic plates. This was clearly the time to take stock and plan for the future.