We report the discovery of a new probable gravitational lens system, associated with the quasar UM 425 = 1120+019, at z = 1.465. The optical image of the system consists of 4 or more components. The two well-observed brightest components have very similar spectra. The difference of the two spectra (scaled appropriately) is consistent with a spectrum of a galaxy at z ≃ 0.6. There is a large number of faint galaxies in the field, suggestive of a lensing cluster at that redshift.

Direct observational information on the structures of the central objects of active galaxies and quasi-stellar objects is hard to come by. Most of what we believe about these regions is inferred from data concerning the emission line spectra, the spectral energy distributions of the featureless continuum, and the luminosities and variability of the active nucleus. Data on the polarization of light from active galaxies, especially spectropolarimetric data which distinguishes between lines and continua, can offer additional insight into the nature of the central object, to some extent more direct information insofar as it can provide geometrical as well as physical information.

The “intermediate zone” is the several decades in radius between the broad and narrow emission line regions in active galactic nuclei. As Antonucci and Miller (1985) first discovered, it comprises two basic structures: an obscuring torus, and a reflection region which fills the hole of the torus and extends some ways above its top. Recent theoretical work, reviewed here, has concentrated on understanding both the thermal/ionization state, and the dynamical state of both regions. The torus is probably composed of a large number of dusty, molecular clouds of large column density and unusually high temperature and ionization fraction, while the reflection region is hot and highly ionized. Spectroscopic diagnostics exist for both. The key problem, still essentially unsolved, in the dynamics of the torus is its geometrical thickness. The reflection region is probably a pressure-driven transonic wind whose source is material evaporated by photoionization from the inner surface of the torus, but a small fraction of it may actually be accreting onto the nucleus.

Emission in the Ca II infrared triplet lines has been detected in 15 AGN. Correlations with optical Fe II emission and O I λ8446 linewidths indicate that Ca II emission comes from neutral gas within the BLR. A new series of ionization models that extend to column densities NH of 1025 cm−2 match the Ca II line strengths and various line ratios. The solution to the Ca II problem, which also solves the “Fe II problem” and the “energy puzzle”, is that a small amount of H− dominates the heating, even for NH ∼1022 cm−2, by absorbing the powerful near-infrared (1 μm) continuum.

The well-known correlations of forbidden lines FWHM with the critical density for collisional de-excitation and with the ionization potential mean, that:

1. i. the noticeable part of forbidden lines emission must arise in a gas of a rather high density n ∼ nc

2. ii. the gas velocity, the density and the degree of ionization they all increase toward the centre of the nucleus.

CCDs make the near-infrared spectral region out to about 1.1 μ observable. A good deal of information on the spectra of AGNs in this range has been published, which we are trying to systematize and extend at Lick. Our original lens-grism spectrograph camera is opaque over the region λ8600-λ10250, no doubt as a result of a coating on one of the elements of the Nikon lens, making this program impossible with it. However our newer “UV-Schmidt” camera is quite effective in this region. With it we have obtained well exposed spectra at moderate resolution (FWHM ≈ 7 Å) of NGC 4151 in three overlapping segments covering the region 7000–11000 Å. Also, for comparison we obtained similar spectra of NGC 1976, the Orion Nebula. The wavelengths and relative line fluxes were measured, and most of the lines were identified. The strongest three lines in both objects are, in order, [S III] λ9532, He I λ10830, and [S III] λ9069. The Orion Nebula spectrum is very helpful for identifying lines in NGC 4151, and for comparison with it. Among the lines identified in NGC 1976, many of them previously known, are [O II], [S II], [Ar III], [Fe II], [Ni II], [Ni III], [Cl II], [C I], in addition to O I λ8446, many H I Paschen lines and He I.

The Brackett α and γ lines of atomic hydrogen, molecular hydrogen H2 v=1-0 S(1) line and [FeII] 1.644 μm line were observed in type 1 and type 2 Seyferts, LINERS, starburst galaxies, and interacting galaxies. For the observations, the Infrared Spectrometer (IRS) with an array of eight InSb detectors was used on the CTIO 4 m and 1.5 m telescopes. The effective size of the slit is 6.8″x4.3″ on the 4 m and 10.3″x20.7″ on the 1.5 m. The resolution of the IRS was 290 km s−1 at 2 μm.

Otherwise acceptable kinematic and dynamic models for emission line regions may be ruled out if the line-emitting clouds are unstable. We discuss the stability of radially-moving clouds to lateral gas flows. Such flows result from internal overpressures arising from the absorption of continuum radiation.

We contend that quantitative measurements of nuclear narrow emission line strength can strongly constrain models that unify AGNs. The reasons for the importance of narrow-line luminosity LNLR are:

1. a) The lines normally arise via photoionisation by the integrated UV/soft X-ray luminosity LPHOT of the central source. Thus LNLR is directly linked to a physical quantity intimately connected with the central engine but not observable from the ground. For constant covering factor we expect an approximate proportionality between LNLR and LPHOT; this has been confirmed observationally for AGNs by estimating LPHOT from either optical non-stellar luminosity or effective ionisation parameter.

2. b) NLRs are far enough from the photoionising source to avoid the excessive obscuration that appears able to attenuate broad-line and continuum emission. Narrow-lines are radiated isotropically unlike, eg, the radio core which may be Doppler boosted. Their variability timescale of 103–4 years is intermediate between those of LPHOT and any large-scale radio emission.

The evidence for anisotropic continuum emission in some Seyferts, is increasingly convincing. However, the generality of this phenomenon has not yet been determined. I present new observations which suggest that its occurrence is widespread.

It is proposed that the clouds thought responsible for the line emission in AGN are not of uniform density but stratified. Such a stratification may be a result either of their self-gravity or of gas dynamics associated with each cloud (e.g. winds). To fix ideas we assume the latter possibility, we examine the consequences and compare with the observations and the phenomenology of emission line clouds.(For the general properties of these clouds see review by Netzer in this volume). Given the successes of the standard model the reader may wonder why is there any need for revisions. The reasons are given in detail elsewhere (see also Scoville and Norman this volume). In brief these are the drag of clouds though the confining medium, the excessive accretion rates implied by the standard model, the response of the line radiation to changes in the continuum, and the lack of a dynamical mechanism for cloud formation. Finally, in the “standard” picture the narrow line clouds are a distinct population with separate dynamics and origin.

The preliminary results of a high-resolution study of narrow emission-line profiles in bright Seyfert galaxies are reported. Profile substructures such as secondary peaks and shoulders are observed in nearly all the objects. The results of profile comparisons in Mrk 359, Mrk 533, and Mrk 1066 are briefly discussed.

Star formation is responsible at least for a part of radiation coming from central regions of active galaxies. Violent bursts of star formation yield rapidly evolving massive stars, which transform in metal-rich enviroments into so called Warmers / Terlevich et al., 1987 / changing nebular spectra to Seyfert or Liner type.

In the last decade, emission-lines from a large number of active galactic nuclei (AGN) have been observed. Most of the models built to explain the observed narrow emission-lines are based on photoionization. Although these photoionization models account for the observed general features, many points remain unexplained and several authors suggest an additional energy source (Ferland and Mushotzky 1984, Ferland and Osterbrock 1986, Stasinska 1984, Viegas-Aldrovandi and Gruenwald 1988). Another possible explanation is suggested by the emitting cloud motions, which account for the observed line widths. If the clouds are moving throughout a dilute gas (n0 ≃ 300 cm−3), a shock can develop. Then, the physical conditions in the cloud are determined by the coupled effect of photoionization and shock hydrodynamics.

In recent years, a large number of high resolution observations of active galactic nuclei (AGN) have become available. Most of them show the [OIII]5007 line profile (Heckman, Miley and Green 1984, Vrtilek and Carleton 1985, Whittle 1985a,b).

Photoionization model calculations suggest the possibility that the coronal line region in Seyfert galaxies may be the result of an interstellar medium exposed to and subsequently photoionized by a “bare” Seyfert nucleus. A series of models are presented in which a “generic” AGN continuum illuminates the low-density phase of the ISM of a spiral galaxy (Korista and Ferland 1988).

Photopolarimetric observations in UBVRI bands of Seyfert galaxies NGC 2992, NGC 3081, NGC 3227 and IC 4329A were carried between 1984–87. Measurements were made with PRL polarimeter (Deshpande et al. 1985) on 1 meter telescope of Indian Institute of Astrophysics, Kavalur. Method of observations and reduction is same as discussed elsewhere (Joshi et al. 1987).

We have made a polarimetric survey of 84 quasi-stellar objects, to supplement published polarization measurements for two samples of quasars identified with flat-spectrum (core-dominant) radio sources in 5 GHz surveys made at Bonn (Kühr 1980) and at the VLA (Perley 1982). The observations were made using the McDonald 2.1m Struve reflector and the polarimeter described by Breger (1979); the typical accuracy is 0.5% for an 18 mag object after half an hour. Earlier surveys, such as those by Stockman, Moore and Angel (1984), included objects of various radio spectral types, and only a small fraction of the objects showed high polarization (> 3%), but our sample of flat-spectrum quasars reveals many more (about half) of the objects to be highly polarized. Some of them are, expectedly, of the BL Lac class, but many of them have strong broad emission lines. There are two striking correlations among the results:

1. (1) The degree of polarization is strongly correlated with the dominance of the radio core - specifically, with the ratio, R, of core to lobe luminosity (Fig. 1). For example, about 75% of the objects with log R > 1.25 and redshift z < 1 have p > 3%. This relation implies that if the radio core radiation is beamed, as seems likely, then so is the optical synchrotron component.

2. (2) The fraction of objects with p > 3% is inversely correlated with redshift (e.g. Fig. 2). The most likely interpretation of this result is that quasars' degree of polarization decreases with decreasing rest wavelength, and the shorter wavelengths are shifted into our wide observational passband at higher redshifts.

We have found that 4 new, bright IRAS quasars, out of 7 observed, have strong, non-variable, wavelength-dependent polarization. Three show degrees of polarization, pλ, increasing from infrared to UV wavelengths (Fig. 1), which implies a combination of a polarized, scattered spectrum and a much redder, unpolarized spectrum. Detailed IR and optical polarimetry and spectrophotometry of one, IRAS 13349+2438 (Wills et al.), shows a polarized flux spectrum, pλxFλ, (continuum and Pa α, Hα, and Hβ broad hydrogen lines) typical of unreddened, luminous quasars. This suggests that the path of scattered light from a central, luminous quasar is low in dust and that the polarization of the scattered spectrum is wavelength independent. The latter is most easily explained by electron scattering although the data do not exclude dust scattering. When this polarized flux spectrum is subtracted from the total spectrum, we are left with a very reddened line and continuum spectrum, E(B-V) = 0.3 to 0.7, which we attribute to the same luminous quasar seen through a thick dusty torus. The angle of polarization is parallel to the major axis of the r-band image, presumed to be that of the host galaxy. If the torus is in the plane of the galaxy, the axial ratio suggests a viewing angle of 40° to the plane of the torus. Fig. 2 illustrates the geometry. The appearance of the quasar at optical and UV wavelengths will depend strongly on viewing angle, suggesting that present samples of quasars selected by colours, optical flux density, or quasi-stellar appearance, may be seriously biased, with important consequences for studies of the space density and evolution of AGN.

At a density of 1011cm−3 or higher, the column emissivity of many lines, such as hydrogen Balmer lines, may not continue to increase with density because the strong radiative transfer results in a “skin effect” near the illuminated front. The Lyα to Balmer line ratio is larger than at lower density.