The fundamental properties of the spacetime fabric of the local Universe cannot be determined without an observable, ultra-accurate standard of inertiality, which can be applied to any celestial object from high-z quasars to an exoplanet orbiting a nearby star. The only currently plausible method of setting such a standard is the establishment of an inertial reference frame based on grid stars anchored by extragalactic sources. An astronomical reference frame at 1 μas accuracy covering the entire sky will make it possible to observe and measure (rather than speculate) a wide range of physical and cosmological phenomena, essentially promoting new observational science. To name a few, it will allow for the first time the direct detection of the acceleration of the Solar System within the Milky Way and the Local Group toward the Virgo cluster. Meaningful experimental constraints will be put on the broadband power of the relic gravitational waves, to constrain the mass function of free-floating planets and comets, to resolve the puzzles of the origin of radiation from radio-loud quasars and the coronal activity in tight RS CVn binaries, and obtain more accurate tests of alternative gravitation theories. All of this will be accomplished by observing the motion and relative position of a large number of stars and distant "fixed" quasars.

We observed the nuclear region of NGC 4261 (3C270) with the VLBA to determine the morphology of the central radio source on parsec scales. Our highest angular resolution image at 8.4 GHz shows a very narrow gap in emission just east of the radio core (on the counterjet side), which we interpret as an absorption feature caused by a small, dense inner accretion disk whose width is less than 0.1 parsec. If the inclination of this inner disk is close to that of the much larger-scale disk imaged by HST, it becomes optically thin to 8.4 GHz radiation at a deprojected radius of about 0.8 pc. September 1997 VLBA observations at higher frequencies should allow us to determine the radial electron density distribution of the inner disk.

We present 5 GHz global VLBI observations of the two Fanaroff-Riley type I radio galaxies B2 0836 + 29 and 3C 465 (2335 + 26). For 3C 465 we present also 1.7 GHz and 8.4 GHz global VLBI data. In addition VLA observations were used to obtain arcsecond-resolution continuum and polarization maps at 5 GHz.Both sources are very asymmetric on the parsec scale, with a core and a one-sided jet, aligned with the main arcsecond-scale jet. We place a limit on the milliarcsecond jet-to-counterjet brightness ratio B-jet/B-ejet greater than or similar to 20 and greater than or similar to 30 for B2 0836 + 29 and 3C 465, respectively. For 3C 465 the strong asymmetry holds to the kiloparsec scale.The brightness asymmetry and the ratio between the core radio power and total radio power allow us to constrain the jet velocity close to the core and the orientation of the radio structure with respect to the line of sight. The results suggest that the plasma speed is relativistic on the parsec scale for both sources, i.e., v(jet) greater than or similar to 0.75c for B2 0836 + 29 and v(jet) greater than or similar to 0.6c for 3C 465. While v(jet) decreases from the parsec to the kiloparsec scale in B2 0836 + 29, in 3C 465 the very high v(jet) holds all the way to the kiloparsec bright spot.Our results are in agreement with the unification scheme suggestion that FR I radio galaxies are the unbeamed population of BL Lac objects. Furthermore, they reinforce the idea that the central engine in FR I and FR II radio galaxies must be qualitatively similar. The different radio morphology could then be due either to an intrinsically different nuclear power, which affects the torus geometry, or to different conditions in the region beyond the parsec scale, where a significant deceleration in the FR I jets occurs.

We present multifrequency radio observations of the low-luminosity (Fanaroff-Riley type 1) radio galaxy NGC 315 at parsec resolution. The observations were obtained at 1.6, 5, and 8.4 GHz with the European and US VLBI networks. In addition VLA observations were used to obtain arcsec resolution maps at the same frequencies.The radio morphology of NGC 315 is asymmetric at milliarcsec resolution. It shows nuclear emission with a core size of 0.05 pc, and a one-sided jet. We place a limit on the milliarcsec jet to counterjet brightness ratio of B(jet)/B(cjet) greater-than-or-equal-to 50 at 1.6 GHz. At arcsec resolution we see a counterjet which is shorter and much more diffuse than the main jet.We derive a spectral index profile at parsec resolution, and estimate the physical conditions in the parsec-scale jet.If we assume that the parsec-scale asymmetry is caused by Doppler boosting of an intrinsically identical jet-counterjet pair, the derived brightness ratio gives beta cos theta greater than or similar to 0.54, which gives an angle theta between the source axis and the line of sight less than or similar to 58-degrees. However, Doppler boosting cannot account for the asymmetry observed on the arcsec scale, given the much lower velocity flow estimated for the arcsec jet. This, together with the lack of any visible motion of the features in the maps at different epochs, suggests that an intrinsic asymmetry in the nuclear engine of NGC 315 is likely to be responsible for the observed radio asymmetry.

We present VLBI and VLA observations of the Fanaroff-Riley ''type I'' radio galaxy 3C 338. On both milliarcsec and arcsecond scales, the source is characterized by emission on both sides of the core. In the VLBI images, the symmetric structures have three or four ''knots'' in the emission on each side of the core. The reliability of the VLBI images was extensively tested using two different data reduction procedures.Detection of equally bright emission on both sides of a parcsec-scale core implies that Doppler boosting is small: either the source orientation is near the plane of the sky, or the twin jets are not highly relativistic. Another possibility is that the jet is intrinsically one-sided and alternately feeds energy into opposite directions.

VLBI observations of the nucleus of Centaurus A have been made at three southern hemisphere observatories. Since Centaurus A is the nearest active galaxy, VLBI investigations are important because the physical processes in the nucleus can be studied in greater linear detail than in other similar galaxies. Previous VLBI observations of Centaurus A have been hampered by its southerly declination (−43°) and the sparsity of VLBI capability in the southern hemisphere, leading to only scattered single point u, v coverage. This paper presents results from the early stages of development of a southern hemisphere VLBI network.