Recommendations of Commission 8 at previous IAU General Assemblies on the following programmes are recalled: Bright Stars (BS), Double Stars unsuitable for photographic observations (DS), Latitude Stars (LS) and Zodiacal Stars (ZS). One more programme – ‘Additional Reference Stars in the areas with Galaxies of the Pulkovo Programme’ (PS) containing 735 stars for the whole sky is proposed.

The Gill Transit Circle at the Cape Observatory has observed zones −40° −52° and −30° −40° of SRS and is now observing zone −52° −64°, which is 80% complete. In addition to fundamental stars selected to define the FK4 system in the zone of observation, clock stars and azimuth stars are also observed so that the observations can be subsequently referred to a more independent or fundamental system.

Since the last report (Clube, 1970), the Cape fourfold overlapping survey of the southern hemisphere has been completed. Approximately one half of the ~6000 plates have now been transferred to Herstmonceux for measurement. The measurement procedure is two staged: initially some 300 of the brightest stars on each of one non-overlapping set of these plates are picked out visually and paper tape listings of rough (x, y) co-ordinates are produced with a D-Mac machine. After some computer editing, these tapes are used to generate search tapes for measuring all the plates on a GALAXY machine. Some 1100 plates have now been measured this way, but the advance to GALAXY measurement has not yet been made on any significant scale. This is because of two difficulties which it is hoped to resolve in the near future, namely the existence of a small astrometric magnitude equation in GALAXY, and the presence on some of the plates of small glass splinters which can damage the ‘nose cone’ of the measurement optics as the plate is tracked beneath it. The technique of reduction and analysis is not described here, but it may be mentioned that the anticipated relative positional accuracy of ~0.″05 is being achieved over fairly wide areas of the sky.

A brief summary is given of the plate materials and results obtained at Yale since the advent of photographic astrometry. The distributions of plates not yet measured or reduced are indicated and possible supplementary programmes for their effective utilization are discussed. Interest centres chiefly in the following areas:

(1) The unmeasured parallax plates.

(2) Reduction of the −60° to −70° zone catalogue.

(3) The El Leoncito programmes for the determination of both positions and proper motions relative to faint galaxies.

(4) The feasibility of a search for intrinsically faint nearby stars with the Yale 40-in. at Cerro Tololo and the 20-in. double astrograph at El Leoncito.

The current astrometric programmes of zone photography and minor planet observations at the Sydney Observatory are reviewed. Future instrumental requirements and resources needed for zone catalogue work are discussed.

The basic techniques of conventional radio interferometry are explained and a review given of the achievements and limitations of the methods as applied to radio astrometry.

The positions of extragalactic radio sources have been determined by very-long-baseline interferometry with uncertainties smaller than 0″.1, and fundamental limits on the accuracy of VLBI astrometry have not yet been reached. Technical improvements within the next five years may enable positions to be determined by VLBI with uncertainties of 0.″005.

Compact components of several radio sources were observed in the course of VLBI observations at S-band between NASA-JPL Deep Space Stations located in Australia, South Africa and California, U.S.A., during the southern summer of 1971–72. These stations were equipped with H-maser frequency standards over this period so that the fringe frequency could be determined to better than 1 mHz.

The internal consistency of the position solution could be assessed for sources observed at least three times. This varied with source declination from ± 0″.1 at a declination of 45° to ± 2″.0 near the equator.

The spread in the position solution for P0727-11, a rapid variable at the observing frequency, was also large.

Positions obtained for some other sources have been used to assist in the identification of associated optical objects.

Inherent in Very Long Baseline Interferometry (VLBI) is the potentiality of determining relative angular positions of radio sources to an accuracy which is superior to present methods by at least one order of magnitude. For VLBI observations such as time delay, fringe frequency and fringe phase the sensitivity to the uncertainties of astronomical constants and geophysical parameters, i.e. luni-solar precession, nutation constant, rate of change of obliquity, polar motion and Earth rotation, is investigated.

The partial derivatives of the observations with respect to these unknowns form the core of the estimation process and give evidence of the influence of the source position and the baseline geometry on their magnitudes. By means of a variance-covariance analysis the standard deviations and the cross correlation coefficients of the unknowns are derived. It is inferred from them that VLBI observations are not only sensitive to variations in the astronomical constants and geophysical parameters, but that the effects caused by these variations are also measurable and separable, if the observations are extended over a sufficiently long interval of time, and the baselines and source positions are chosen appropriately.

Radio positional measurements have achieved an accuracy as high as that of optical astrometry, with uncertainties no greater than a few hundredths of an arc second in each coordinate. Declinations and relative right ascensions are determined absolutely. Since the principal sources of systematic error are different for radio and optical astrometry, radio measurements can be useful in the preparation of future fundamental catalogues.

The 5-element Texas Interferometer is engaged in a 365 MHz survey of the sky with a primary goal of establishing positions of about 50000 discrete radio sources with an accuracy of about 1″ in each coordinate. Measurements are made relative to optical positions of identified sources, and the 2000 positions thus far obtained support our expectations of the accuracy of the survey. In a companion program, optical positions of objects near radio source positions are being measured to ±1/2″ accuracy on glass copies of the Palomar Sky Survey, yielding both improved overall calibration of the radio positions and identification of associated optical objects on the basis of position coincidence alone, without the selection effects usually introduced by auxiliary identification criteria.

Measurements by C. M. Wade and R. M. Hjellming with the three-element-interferometer at Green Bank were reduced simultaneously for instrumental parameters and positions of 59 radio sources. The minor axes of the confidence ellipses have typical values of about 0″.15 while the major axes vary with declination from 0″.15 to 1″.4.

The use of radio interferometric techniques to monitor geophysical phenomena such as UTI-UTC variations, polar motion, and tectonic plate motion requires that there be available a fairly uniform distribution of extragalactic radio sources whose relative positions are known to better than 0″.01. These sources must also have very small proper motion, and not be subject to asymmetric changes in radio structure or to very large variations in intensity. In addition, to facilitate tying this frame of reference to the optical star frames, it is desirable that many of these radio objects have optical counterparts. This paper describes our progress in obtaining such a frame and sets forth our desiderata for further work by radio and optical astronomers.

Proper motion work at the present time relies on the availability of plates taken in the past, for which the magnitude limit is generally not fainter than about mpg = 16, apart from Luyten's survey with the Palomar Schmidt. It is very important for future studies to fainter magnitudes in, for example, star clusters, that sufficient time is made available on large telescopes so that plates can be taken now. Consideration should also be given to an extension of proper motion work in Selected Areas.

Ideally, trigonometric parallax observations are best carried out on a dedicated telescope, such as the USNO 61-in. reflector, with a large amount of time available; but if sufficient observational material is to be obtained on faint stars, time must also be taken on other medium sized and large reflectors. With the severe limitation of time available on these telescopes, the criteria for selection of stars for such programmes become of prime importance. Large Schmidt telescopes, with fast automatic plate measurement, may possibly be used for surveys of nearby stars, unbiased by any kinematic or photometric selection effects.

The U.S. Naval Observatory program on trigonometric stellar parallaxes with the 61-in. astrometric reflector has been in progress since 1964. To date 201 definitive and 8 preliminary negligible parallaxes have been published, including the UBV photometry of these stars. Data for an additional 35 stars are still unpublished.

The mean error in position for an image of unit weight is 1.2 μ or 0″.016, of which 0.8 μ originates from the measuring error of the automatic measuring machine.

An error of the parallax of 0″.004 can usually be obtained with 40 or fewer plates and with an average parallax factor of 0.7.

A statistical investigation of the derived parallaxes shows that they are free from significant internal systematic errors.

In 1971 a pilot parallax programme was initiated on the 98-in. Isaac Newton Telescope at the Royal Greenwich Observatory. Nineteen stars between visual magnitude 15 and 19 and of high proper motion were put on the programme. The plates were taken at prime focus through the Wynne corrector. Observations are now complete for two of these stars. Because of time allocations and the desire to observe at maximum parallax factor, observations are made at large hour angles. All stars in each field are being measured on the GALAXY measuring engine and parallaxes and proper motions will be computed for all stars. The fields are being reduced using an overlap reduction technique.

The present state of the Palomar Proper Motion Survey is described. Before the CDC automated-computerized plate scanner became operational in 1972, 153 plate pairs had been blinked, whereas since then 443 pairs have been machine processed. Of the remaining 350 fields, some 40 are probably beyond the capability of the present system due to excessive star density.

Some 21000 new motions larger than 0″.18 annually have been published. The accuracy of an individual motion is about ± 0″.016 from a single machine-processed plate pair, as compared with ± 0″.025 from hand measured plates.

There is an indication that the number of large motions per unit area increases with galactic latitude.

Optical positions measured directly on the prints and plates of the Palomar Sky Atlas have been compared with various published lists of precise optical positions. The external comparisons suggest that positional errors ~ ±0″.3 in each coordinate have been achieved. This supports the claimed accuracy and establishes the Sky Atlas as an important astrometric medium for investigating radio-optical associations.

A method is proposed for improving the astrometric accuracy obtainable from curved-field Schmidt plates. A reseau of small crosses in a 10 mm-spaced rectangular grid is imposed on the plate while still curved in its holder. The reseau is calibrated by photography of a field of reference stars with known positions. The method eliminates the troublesome errors arising from uncertainty in the nature of deformation of the curved plates.