The principal purpose of the IAU system of astronomical constants is to provide a self-consistent set of constants for use in the computation of the international ephemerides of the Sun, Moon, planets and stars and in the reduction of observations of these bodies. At present nutation is computed from a theory of the rotation of the Earth as a rigid body and only the coefficient of the principal term in obliquity is specified in the system of constants. Such a simple specification will not be adequate for use with the more precise observations that are becoming available, and it appears that it will be necessary to adopt a new model of the Earth and to develop a new theory of nutation which will take into account the elastic properties of the Earth. The new model should be consistent with other constants of the IAU system, and with the model used in other branches of geophysics. The new specification of nutation should be formally adopted by the IAU in 1979 so that it can be used in the published ephemerides for 1984 onwards.

Fedorov (1958) made a splendid discussion of data on the nutations; Vicente and I made a theory, taking account of a fluid core and of an elastic shell. The agreement was far better than had been found before, but some inconsistencies remained.

Analyses of the nutation terms based on the results of astronomical observations of time and latitude provide valuable information to promote studies on the constitution of the Earth's interior. Indeed, the results of analyses of the nutations agree well with those predicted by theory for a fluid core. This shows that the matter of the outer core, 3,000 km and more below the Earth's surface, is not only in a fused state but also is a “fluid” which can move in response to the external forces. Many additional studies on the nutation terms are now in progress, for example, to estimate an extreme value of viscosity of the fluid core, to discuss the possibility of the geomagnetic dynamo being driven by the precession, and so on.

The useless introduction of the new words “wobble” and “sway” has produced a flourishing rise in confusing terminology in most of the recent papers. These two words even if picturesque are fundamentally unprecise. They perhaps help to describe – in the English language – the nature of the movement but in no way correspond to the classical terminology of fundamental mechanics. They have apparently no convenient translation into the other usual languages.

Two fundamental frames of reference are used in the study of the rotation of the Earth: the nonrotating celestial coordinate system XYZ attached to the directions to stars and/or extragalactic sources, and the terrestrial coordinate system xyz attached in “a prescribed way” to several points (observatories) on the surface of the Earth or to the pencil of unit vectors drawn from an arbitrary origin parallel to the local verticals at these points.

The scale values used in the work of the International Latitude Service are known to be in some cases evidently erroneous. After trying various methods to improve these values, E. P. Fedorov (1963) decided on the one based on the computation of the mean latitudes obtained from observations of the pairs with zenith distances of opposite sign.

The non-polar common terms, z and τ, of the IPMS system are analyzed for determining amplitudes of some terms of the forced nutation. The method of analysis and the results obtained are described briefly. Comparison is made with theoretical and other observational results. A fuller account of this work will be published elsewhere.

There are two reasons why the study of fortnightly nutation is important. First, it provides an opportunity to investigate some peculiarities of the Earth's structure; second, it permits us to calculate more precisely the apparent locations of stars.

Since 1953 regular observations have been made with the polar photographic tube at Pulkovo, a fixed astrograph (D = 20 cm, F = 600 cm), which can take photographs of a circumpolar zone of 60′ radius (Bakhrakh, 1971). An analysis of measured polar distances of the same 30 stars on each plate gives corrections to the adopted values of the constants of aberration and precession and of the coefficients of nutation. At present the determination of the coefficients of the short-period semi-annual solar and fortnightly nutations is of greatest interest, as they are affected most by the Earth's liquid core.

The effects of some short-period tidal and nutational terms in the observations of the Herstmonceux PZT have been re-examined using observations that extend over a longer period and have been re-reduced using improved catalogue positions. Analytic smoothing of these data has furnished observational residuals that reveal very clearly the effects of short period terms with relatively small amplitudes.

The first attempts to take into account the dynamic effect of a liquid core on the rotation of the Earth were made as far back as the end of the 19th century.

The observations of bright zenith stars (α Persei and η Ursae Maj) have been carried out in Poltava since 1939. These observations have been used for determining the corrections to semiannual and fortnightly nutation terms and for studying the nearly diurnal free polar motion (Popov, 1968; Popov and Yatskiv, 1977). Recently the reduction of the new series of bright zenith star observations has been finished in Poltava (see Table 1). Therefore it was interesting to carry out a new analysis of these observations for the purpose of studying the nutation and of searching for the nearly diurnal free polar motion.

Astronomical latitude observations of the ILS stations seem to indicate the existence of both a retrograde and a direct component of nearly diurnal polar motion with close frequencies (Yatskiv et al. 1975). The latter would manifest itself as a latitude variation with a period of about 204 mean days (m.d.) and with a phase depending on the double longitude of station.

Le travail entrepris ici porte sur les mèmes observations de latitude effectuées à Paris, au moyen d'astrolabes, que celles utilisées dans d'autres études (Capitaine, Chollet et Débarbat, 1977; Capitaine, 1977). Pour des raisons qui apparaitront plus loin nous n'avons pu utiliser que la période comprise entre septembre 1970 et le début de 1977.

Any error in the position of the Earth's axis of rotation in space gives an error in the equatorial coordinates of stars used in the reduction of latitude and time observations and consequently gives an error in the results of these observations. This may be due to an error in the classical terms of nutation or to the neglected nutation in space produced by the theoretically predicted nearly-diurnal wobble (Hough, 1895; Poincaré, 1910; Jeffreys, 1949; Molodensky, 1961; Toomre, 1974). A preliminary study of these errors in latitude and time observations was reported in a preceding paper (Capitaine, 1975), which was subsequently corrected for a minor sign error (Capitaine, 1977).

Les données utilisées dans ce travail sont les observations de latitude couvrant une durée de vingt années (1956 a 1976). Le programme d'observation est resté inchangé et les réductions ont toutes été effectuées dans le Système du FK4 et des Constantes adoptée par l'UAI en 1968.

The method for deriving the amplitudes of the principal term of nutation from the z-term of latitude and the corresponding term of UT (here denoted by w) is well known (Yokoyama, 1977).