1. INTRODUCTION: THE INNER STRUCTURE OF AGNs

There is now general consensus that the long-standing paradigm for active galactic nuclei (AGNs) is basically correct, i.e., that AGNs are fundamentally powered by gravitational accretion onto supermassive collapsed objects. Details of the inner structure of AGNs, however, remain sketchy, although both emission lines and absorption lines reveal the presence of large-scale gas flows on scales of hundreds to thousands of gravitational radii. The accretion disk produces a time-variable high-energy continuum that ionizes and heats this nuclear gas, and the broad emission-line fluxes respond to the changes in the illuminating flux from the continuum source. The geometry and kinematics of the broad-line region (BLR), and fundamentally its role in the accretion process, are not understood. Immediate prospects for understanding this key element of AGN structure do not seem especially promising with the realization that the angular size of the nuclear regions projects to only microarcsecond scales even in the case of the nearest AGNs. Unfortunately, there is only very limited information about the BLR from the emission-line profiles alone, since many simple kinematic models are highly degenerate. Nevertheless, it has been possible to draw a few basic interferences about the nature of the BLR:

1. There is strong evidence for a disk component in at least some AGNs. In particular, there is a relatively small subset of AGNs whose spectra show double-peaked Balmer-line profiles. Double-peaked profiles are generally associated with rotating Keplerian disks.

2. There is strong evidence for an outflowing component in many AGNs. Some emission lines have strong blueward asymmetries, suggesting that we preferentially observe outflowing material on the nearer side of an AGN. Slightly blueshifted (relative to the systemic redshift of the host galaxy) absorption features are quite common in AGNs, and there is a good deal of evidence that this absorption, seen primarily in ultraviolet and X-ray spectra, arises on scales similar to that of the broad emission lines.

3. There is strong evidence that gravitational acceleration by the central source is important. As discussed below, a physical scale for the size of the line-emitting region can be obtained by the process of reverberation mapping. The derived scale length for each line is different, with lines that are characteristic of high-ionization gas arising closer to the central source than lines that are more characteristic of low-ionization gas, thus demonstrating ionization stratification within the BLR. Moreover, the higher ionization lines are broader, and indeed the relationship between size and velocity dispersion of the line-emitting region shows a virial-like relationship, i.e., r V^-2, where r is the characteristic scale for a line which has Doppler width V.

The conclusion that gravity is important leads us directly to an estimate of the black hole mass, which we take to be

(1)

where G is the gravitational constant and f is a scaling factor of order unity that depends on the presently unknown geometry and kinematics of the BLR.

In this brief introduction, we already see the two major reasons that understanding the BLR is of critical importance to understanding the entire quasar phenomenon: (1) we need to understand how the accretion/outflow processes work in AGNs and (2) we need to understand the geometry and kinematics of the BLR to assess possible systematic uncertainties in AGN black-hole mass measurements.