We observed a total of 13 low-redshift AGN with the Hopkins Ultraviolet Telescope (HUT) during the course of the Astro-1 and Astro-2 missions. Of these, 4 show intrinsic Lyman-limit absorption — NGC 1068, NGC 4151, NGC 3516 and NGC 3227. AU galaxies with optically thick Lyman limits have extended narrow-line regions with bi-polar morphologies. All AGN with no absorption are Seyfert Is with compact NLRs. These observations support geometrical shadowing as the means for collimating the ionizing radiation in unified models of AGN, most likely in a photoionized atmosphere above the obscuring torus.

Simple unified models predict that many Seyfert 2s should be highly polarized ( ≳ 50%), but such highly polarized Seyfert 2s are not observed. As one of the possibilities to explain this problem, the existence of clumpy scattering clouds with optically thick condensations in the nuclear regions is discussed.

A global view is given of how emission lines have been, and may in the future, be used to enhance our understanding of AGN. Lines from the microwave to the X-ray bands all contribute. Although we have a deep understanding of the physical processes by which line photons are generated, when the circumstances of line emission are complicated, models become too unreliable to provide strong inferences about the rest of the AGN system. At present, the lines which appear the most promising for helping answer the most important questions about AGN are the 22 GHz H2O rotational transition and the 6.4keV FeKα fluorescence.

We have studied the dependence of X-ray spectral and luminosity properties on radio morphology using ROSAT data, for a sample of radio quasars. The results support the unified schemes for FS and SS quasars.

LINERs are found in ~30% of all bright galaxies, including luminous infrared galaxies. They form a heterogeneous class powered by a variety of ionizing mechanisms such as low-luminosity AGNs, starbursts, shocks, or any combination of these.

In early-type spirals, LINERs are powered by a low-luminosity AGN, or by an AGN surrounded by circumnuclear star-forming regions. In luminous infrared galaxies, LINERs are powered by starbursts with associated wind-related extended shocks, and an AGN may play a minor role, if any. LINERs in some FR I radio galaxies show a strong evidence for the presence of a massive central black hole, and there are indications for the existence of shocks in the nuclear disks of these galaxies. Yet, the dominant ionizing mechanism for LINERs in radio-quiet ellipticals and FR I host galaxies is still unclear.

Multifrequency high spatial resolution imaging and spectroscopy are essential to discriminate among the different ionizing mechanisms present in LINERs.

We summarize the current state of our project to determine the gravitational potential in the innermost few hundred parsecs of Seyfert cores from observed line-of-sight velocities of molecular clouds. Therefore we distinguish between the observational part on the one hand and the theoretical aspects on the other. On the theoretical side the problem is lack of knowledge of the flow pattern and the spatial distribution of the clouds. We analyze the possibility of distinguishing between the different flow patterns and determining the mass distribution. On the observational side the problem is obtaining sufficiently high spatial and spectral resolution.

We have investigated the characteristics of AGN using data from ASCA X-ray observations and considered their contribution to the CXB. There are mainly four specific topics that need to be considered: (1) the absorption edges due to warm gas, (2) the complex spectra of Seyfert 2 galaxies, (3) the broad iron emission line, and (4) the soft X-ray excess.

The CXB can be explained by superposing AGN spectra according to cosmological distance, but so far the distant quasars (z > 1) that have been observed show spectra in the 2–10 keV band that are steeper than the spectrum of the CXB. Recent observations have been more clearly defining the properties of the AGN one by one. We have more than 80 high-quality AGN spectra, but their properties are not well-balanced due to poor sampling of low-luminosity AGN. A large number of lowluminosity AGN might be the main contributors to the CXB.

I briefly outline some of the unanswered questions that have emerged as recurring themes at this conference and offer some speculation about how these will be addressed with new techniques and facilities of the future.