A short study of FK4 catalogue is based on CGA (Billaud G. et al, 1978). Important differences are shown in right ascension as well as in declination and assumptions are made to explain the discrepancies between both catalogues.

IAU Symposium No. 78 “Nutation and the Rotation of the Earth” held in Kiev in 1977 revealed a certain lack of precision in the fundamental concepts and some looseness of terminology employed in the treatment of this problem. When talking about polar motion we should give, first of all, rigorous conceptual definitions of both the pole and a reference frame in which it moves. The selection of a reference system was the topic of an IAU Colloquium held in Torun in 1974. Although the discussion there was thorough and comprehensive, it did not result in the removal of all ambiguities which have tarnished discussion of the problems in the understanding of the Earth's rotation.

The recomputation of the past ILS observations has been carried out at Mizusawa and partly at Cagliari. Preliminary reduction of the observations at the northern stations has been completed. The coordinates of the pole were calculated preliminarily and were compared with those by Vicente and Yumi (1969, 1970). It is known that the coordinates of the pole in the past ILS reports require considerable corrections which are probably mainly due to errors of the micrometer values. Magnetic tapes of the original observational records and of the individual latitudes are now available on request.

Improved series of Universal Time and latitude measurements back to 1962 have been provided to the BIH by several observatories recently. New techniques are currently or will soon provide Earth rotation data that are independent from the classical astronomical observations. In the meantime, the BIH has acquired experience on possible methods for achieving better accuracy. These reasons make it worthwhile to apply our present practical knowledge to the past data. The method which will be used for computing Earth rotation data from the updated BIH files is presented.

The motion of the Earth's axis of inertia has been derived, taking elastic deformation into account, from the polar coordinates determined by the BIH for the period from 1962.0 to 1975.0. Characteristics of the motion of both the pole of inertia and the pole of rotation have been examined. The secular displacement of these poles relative to the pole defined by the low order harmonics C21, S21 determined from observations of satellites seems to confirm that the inertial reference axis has an apparent wandering motion within the deformable Earth.

The mean astronomical latitude of the Tashkent Observatory was determined by D. D. Gedeonov in 1895–1896 before the International Latitude Service had been organized. The determination was based on a continuous fourteen-month series of observations with a Wanschaff visual zenith telescope (d = 68 mm., f = 870 mm.). These data consisted of the observations of 2214 Talcott pairs which were observed from July, 1895 through August, 1896.

The polar orbit was computed in a system of ten stations with small variations of mean latitudes and compared with the polar orbit based on ten stations with large variations of mean latitudes.

Astrolabe observations made at Paris have been analysed to obtain more reliable values of the constants of nutation. A new method of reduction which takes account of variations in the zenith distance of the observations gave values for the 18.6 year and 9.3 year terms which are almost the same as those derived by classical methods, but for the annual and semi-annual terms the results are different. It is proposed to apply the new method to other series of observations.

The National Observatory of Brazil was established in 1827 to study the problems of latitude and time. Time and latitude programs are still an important part of its work, but are now accompanied by increasing activity in the fields of geophysics, astrophysics and radio astronomy.

Results of observations made with the Danjon astrolabe at Merate Observatory in the period from 1970 through 1976 were analyzed. The observational program and methods have been discussed previously (Buffoni et al., 1975a, b). The observational accuracy of the observations was compared with that of the Paris and San Fernando instruments. No substantial difference was found.

As a part of the establishment of a unified longitude system for the European Triangulation Network the difference in longitude between the reference points Merate and Milano was measured with a Danjon Astrolabe. This paper describes the results of a one-step adjustment of these observations including additional parameters for effects like personal equations and catalogue errors.

The results of the spectral analysis of the velocity of the polar motion based on the data of the International Latitude Service (ILS), International Polar Motion Service (IPMS),. Bureau International de l'Heure (BIH), and the Doppler Polar Motion Service (DPMS) are presented.

In order to discuss accurately the motions of the Earth in space, it is necessary to define rigorously two readily accessible reference systems. The conception and the realization of celestial absolute systems and terrestrial coordinate systems are discussed. It is suggested that these systems of reference ought to be defined with a minimum of theoretical or observational constraints. Examples of such ideal reference systems are given, together with some desirable properties for intermediate systems.

The basic problem of fundamental astrometry is to relate the instrumental reference frame of an observer to the frame defined by ephemerides of stars and objects in the Solar System. It is shown that in principle the choice of definition of the Ephemeris Reference Frame (ERF) can be quite arbitrary. For convenience, it is argued that the ERF should be defined by the rotation of an axi-symmetric rigid model Earth, the celestial pole being the direction of the axis of figure. This definition has practical and theoretical advantages over a model-dependent definition which attempts to take account of non-rigidity of the actual Earth.

In order to improve the representation of nutation, the effect of elasticity of the Earth on the nutation in space and diurnal nutation of the terrestrial rotation axis is considered and its amplitude is evaluated for the principal terms. The choice between several methods taking this effect into account is discussed. A comparison with the effect induced on nutation by the existence of a liquid core in the Earth's interior shows that the consideration of elasticity alone cannot give any amelioration in the representation of nutation.

Modern high accuracy measurements of the non-rigid Earth are to be referred to four-dimensional, i.e., time and space-dependent reference frames. Geodynamics phenomena derived from these measurements are to be described in a terrestrial reference frame in which both space and time-like variations can be monitored. Existing conventional terrestrial reference frames (e.g., CIO, BIH) are no longer suitable for such purposes.

The ultimate goal of this study is the establishment of a reference frame, moving with the Earth in some average sense, in which the geometric and dynamic behavior of the Earth can be monitored, and whose motion with respect to inertial space can also be determined.

The study is conducted in two parts. In the first part problems related to reference directions are investigated, while the second part deals with positions, i.e., with reference origins. Only the first part is treated in this paper.

The approach is based on the fact that reference directions at an observation point on the Earth's surface are defined by four fundamental vectors (gravity, Earth rotation, etc.), both space and time variant. These reference directions are interrelated by angular parameters, also derived from the fundamental vectors. The interrelationships between these space and time-variant angular parameters are illustrated in a commutative diagram–tower of triads, which makes the derivation of the various relationships convenient.

In order to determine the above parameters from observations (e.g., laser ranging, VLBI) using least squares adjustment techniques, a model tower of triads is also presented to allow the formation of linear observation equations. Although the model tower is also space and time variant, these variations are described by adopted parameters representing our current knowledge of the Earth. For details, see Bulletin Géodésique, end of 1978.

At this symposium we are to hear a great deal about new techniques for the measurement of earth rotation and polar motion that have come into being in the last decade and I am privileged to give a short introduction to one of these new techniques, that of radio interferometry.

Radio interferometric techniques for the determination of Earth rotation are described.

The National Geodetic Survey (NGS) of the National Ocean Survey (NOS), a component of the National Oceanic and Atmospheric Administration has begun a project to establish and operate a 3-station network of permanent observatories to monitor polar-motion and Earth rotation (UT1) by radio interferometric observations of quasars. The project designation is POLARIS (POLar-motion Analysis by Radio Interferometric Surveying).

The POLARIS observatories will be equipped with a new generation of instrumentation and software, the Mark III data acquisition and processing system currently under development by a multi-organizational team.

The Deep Space Network (DSN) [operated by JPL under contract to the National Aeronautics and Space Administration] is implementing a Very Long Baseline Interferometry (VLBI) capability at DSS 63 (Spain), DSS 14 (California, USA), and DSS 43 (Australia) to support the navigation requirements of planetary space missions. The early development work for this system has already demonstrated the capability of measuring UT1 with a formal accuracy as low as 0.6 msec with only 6 hours of data. Further, a radio astrometric catalog of approximately 45 sources whose positions are known to better than has been constructed. In addition to these measurements, this paper describes the characteristics and anticipated performance of the complete VLBI system being implemented within the DSN for operational use in mid-1979. In particular, one of the capabilities of this system will be the measurement of UT1 and polar motion at weekly intervals. Although the navigation accuracy requirement is only 50 cm for the Voyager mission, this system should be capable of delivering UT1 and polar motion determinations with decimeter accuracy if it is operated at maximum performance. An additional requirement of this operational system is that it have the capability of providing these results within 24 hours of the actual observations.