We have performed exact numerical calculations of the Stokes parameters I, Q, and U for optically and Faraday thick inhomogeneous relativistic jets. We consider how changes in Faraday depth, as well as in the observer-jet geometry, change the resolved and unresolved properties of the jet.

We present a global fringe fitting technique for the polarized fringes in VLBI. The standard search method imposes a signal-to-noise (SNR) limit on usable data. In our method the search procedure is circumvented and the SNR limitation removed.

Polarized radiation of cosmic radio sources carries the important information of the radiation mechanism and magnetic field structure of the sources. The background nonpolarized radiation of extended radio sources may also be partly transformed into spurious polarization. No comprehensive theory concerning the whole physical process of radiation transfer has been developed.

The λ6 cm total intensity and polarization structures of OJ287 have been determined at two epochs a year apart. At each epoch the source can be modeled as a linear structure consisting of a core and two knots at position angle PA ≃ −115°. This structure expanded slightly over a year, each knot increasing its separation from the core by ≃ 0.3 mas, suggesting an apparent superluminal velocity of ≃ 3.3h−1c for each.

New VLA polarization data for 140 radio sources have been used to study the correlation between VLBI structural position angle and radio and optical polarization position angles reported in Rusk and Seaquist (1985). We confirm that there is a strong tendency, in active galaxies and quasars, for the radio E vector (of the core component) to lie normal to the VLBI structural axis. However, in BL Lac objects we find a tendency for the core radio E vector to be aligned parallel to the VLBI structural axis.

The λ6 cm milliarcsecond polarization structure of 3C345 has been determined at three epochs between December 1981 and March 1984. The knots C2, C3, and C4 all showed changes as they moved away from the core, which remained virtually unpolarized.

Radio imaging of VLBI core-jet sources can be used to examine the case for continuity of jet-like features between parsec and circumgalactic scales. Futhermore, polarimetry of such sources allows investigation of the dominant magnetic field topologies as a function of linear offset from the central engine. Examination of these continuity and field topology issues is essential for an understanding of how energy is channeled from the nuclear regions to the circumgalactic environment.

We discuss the λ6 cm total intensity and polarization structures of a number of BL Lacertae objects at milliarcsecond resolution. 0235+164 was unresolved and weakly polarized at each of two epochs a year apart; each of the other objects displays structure in polarized flux. 0735+178 and 1749+096 can be adequately modeled by two or three point components—a “core” plus one or two “knots.” The core components were moderately polarized (≃ 5%), while “knots” may be polarized at 8% or more, consistent with these components being optically thin. Preliminary results for BL Lac indicate that the total intensity structure can be modeled well by a set of four gaussian components; the polarization structure is complex, but is dominated by the northernmost knot in the jet.

Studies of H2O masers have demonstrated the power of VLBI techniques to measure relative positions with sufficient accuracy (∼ 10 μas) to determine proper motions and to estimate distances to maser sources throughout the Galaxy. The distance to four H2O masers have been determined, and the distance to the center of the Galaxy has been determined to be 7.1 ± 1.5 kpc from observations of the H2O masers in Sgr-B2. Proper motion distances for other H2O masers, and possibly for OH masers, may allow the determination of the fundamental parameters describing the size (Ro) and rotation rate (Θo) of the Galaxy with accuracies of better than 10%. Finally, the measurement of the proper motions of H2O masers in nearby galaxies (< 10 Mpc) is feasible and offers the possibility of direct calibration of the extragalactic distance scale.

The most direct method of determining distances to galactic sources outside the Solar System is the measurement of annual parallaxes. Using as an example astrometric observations of the quasars 3C345 and NRAO512 which are separated by ∼ 0°.5 at the sky, we show that under certain conditions, annual parallaxes of galactic radio sources can be determined with VLBI with an accuracy of 50 μas (standard error). This accuracy is about tenfold higher than any annual parallax accuracy ever obtained with optical astrometry.

The, so far, most direct method of determining distances to extragalactic sources combines VLBI measurements of a supernova's angular expansion velocity with optical spectroscopic measurements of the supernova's radial expansion velocity. We have estimated the distance to SN1979C in M100 in the Virgo cluster to be a lower bound on the distance to SN1980K in NGC6946 to be 5 Mpc, and H0 to be with the uncertainties intended to represent standard errors.

VLBI observations at 2.3 GHz of SN1987A on 28 February 1987 yielded no fringes, implying, for an optically thin shell, a lower bound on the (outer) diameter of 1.9 mas. From the comparison of the VLBI and optical results, we infer that the radiosphere of SN1987A was either about equal to, or larger than, the photosphere of the supernova five days after the explosion.

We have studied the brightest of the many compact radio components, designated 41.9+58, in the central region of M82 (Unger et al. 1984; Kronberg et al. 1985) with the EVN at frequencies of 1.6 and 5 GHz (Wilkinson and de Bruyn, 1984; 1987).

A prompt radio burst has been observed from the supernova 1987a in the Large Magellanic Cloud. Observations were made at 0.843, 1.415, 2.29, and 8.41 GHz. At frequencies around 1 GHz, the peak flux density reached about 150 mJy and occurred within four days of the supernova. This event may be a weak precursor to a major radio outburst of the type previously observed in other extragalactic supernovae. Radio monitoring of the supernova is continuing at each of the above frequencies, and coordination is underway of a southern hemisphere VLBI array to map the radio outburst region as it expands. Differential astrometry carried out on prime-focus plates taken with the Anglo-Australian telescope indicates that the component, star 1, of Sanduleak's star SK-69202 is within 0.05 ± 0.13 arcsec of the supernova.

Four specific and particularly powerful types of possible VLBI lens studies are discussed. First, comparison of mas scale structure in putative pairs of images separated by arc seconds can provide a powerful additional test of the lens hypothesis in specific candidate systems. Second, VLBI searches for lens systems with image separations too small for resolution by optical or VLA searches will limit (or even determine!) the cosmological density of condensed objects with individual masses ∼ 106M⊙. Third, study of multiply imaged superluminal expansion events will allow a determination of the light travel time delay between different images in a lens system, a quantity which is quite difficult to measure by other means but which would allow profound cosmological tests. Fourth, VLBI data can be used to determine relative image parities and even the full magnification matrix of various images in a lens system, thus providing powerful additional constraints on detailed lens models. Finally, the speculative possibility of detecting Galactic stellar lensing events using VLBI techniques is discussed.

A series of VLBI observations of the gravitational lens system 0957+561 at λ13 cm has yielded the positions of the A and B images, the relative magnification of their largest discernible radio structures, and the time variability of their smallest discernible radio structures. These observations have also allowed upper limits to be placed on the flux density of an expected third image. The positions and relative magnification of the A and B images provide new information with which to constrain models of the lens that forms the images. The detection of variations in the flux densities of the cores of A and B suggests that observations at shorter wavelengths may reveal superluminal motion, which may in turn provide a means to measure the relative time delay.

We have used the relative positions and magnifications of the A and B images in the gravitational lens system 0957+561, obtained from VLBI observations, to constrain a model for the surface mass distribution of the lens. With measurements of the difference ΔτBA in propagation times associated with A and B (the “relative time delay”) and of the velocity dispersion of the main lensing galaxy, both to be obtained, our model will yield a value for H0 with an uncertainty of ∼ 20% due mainly to uncertainties in our assumptions.

On June 1, 1984 we conducted a seven station 18-cm VLBI observation of the 2016+112 gravitational lens system. Preliminary brightness distributions for A and B have been obtained via model fitting. Weak correlated flux density was detected in the C component region.

VLBI experiments can contribute to our understanding of absorbing gas observed in the spectra of high-redshift quasars by measuring the spatial extent of the absorbers. An optical survey conducted by Wolfe, Turnshek, Smith, and Cohen (1986) has turned up a class of absorbers characterized by large HI column density and a rich spectrum of metal absorption lines. In these respects the absorbers resemble spiral galaxies, but their occurence is too frequent for interception probabilities based on galaxy cross sections at the present epoch. Recent radio observations to determine the nature of such an absorber at z=2.04 in the spectrum of PKS0458-020 include both continuum observations at 606 and 1590 MHz to define the structure of the background quasar as well as spectral line VLBI at 467 MHz. The observations reveal structure in the continuum source on a wide range of angular scales, making the source nearly ideal for spectral line VLBI on a range of baselines. Early spectroscopic results show that the absorber must have a spatial extent of galaxian size, consistent with the hypothesis that these absorbers are galaxies. The further implication is that galaxies must have been larger and richer in gas at the epoch around z=2 in order to provide the likelihood of interception that is observed optically. (A complete description of the result is in preparation by Briggs, Wolfe, Liszt, Davis and Turner.)

A review is presented of recent results from interferometric observations of cosmic masers. Several topics will be discussed, including MERLIN/VLA/VLBI observations of megamasers; recent results on masers in star forming regions and around late-type stars and some new results on the effects of interstellar scattering on maser spot sizes.

The theory of astronomical masers is reviewed. As with laboratory masers, masing occurs when a transition between two energy levels of a molecule exhibits inverted populations. In order to present the basic concepts about masers, an idealized two-level system is used. The exact energy level structure is taken into account later on when the pumping of specific molecules is discussed. Unlike laboratory masers, where the radiation must be bounced between two mirrors to accumulate gain, the propagation of radiation in astronomical masers is a lot simpler. This is because astronomical masers are single-pass and broadband. Thus, the main theoretical effort has concentrated on inventing efficient mechanisms that produce population inversion. Specific pumping mechanisms for the three molecules (H2O, SiO, and OH) that exhibit strong masing are presented and their ability to explain the observations is discussed.