In two earlier papers (Zickgraf et al. 1997a; Appenzeller et al. 1998; hereafter referred to as "Paper I" and "the Catalog", respectively) we described and presented the optical identification of a complete sample of X-ray sources discovered in the ROSAT All-sky Survey (RASS, Voges et al. 1999). A statistical analysis of these identifications (which in the literature are also referred to as the "RASS Selected Area-North survey") has been presented by Krautter et al. (1999). Subsamples of the Catalog have been discussed by Zickgraf et al. (1998b), Mujica et al. (1999), and Appenzeller et al. (2000). The present paper presents a discussion of the Seyfert galaxies and QSOs listed in the Catalog as main optical counterparts of RASS X-ray sources with an Identification Quality Index or (i.e. reliable identification on the basis of our spectra and/or on the basis of reliable literature data). Since the Catalog lists no Seyfert galaxy and only one QSO with ("uncertain identification") the present paper covers practically all objects of these types in our survey.
Not included in the following discussion are the BL Lac objects (which are discussed elsewhere), the two LINERs, and the 7 AGN for which due the weakness of the lines or inadequate no subclass could be assigned in the Catalog. Also omitted are 3 Seyfert galaxies and one QSO which - although inside the error circle of an X-ray position - were not regarded to be the main source of the observed X-ray radiation since another (in our opinion dominant or more likely optical counterpart was (also) present at the position in question.
As described in Paper I our identifications and classifications were based on low-resolution spectra obtained with a multi-object spectrometer at a 2.1-m telescope. These spectra were exposed to reach a suitable for a reliable classification and redshift derivation. Originally there were no plans for a quantitative evaluation of these spectra. However, after photometric calibration the great majority of the spectra turned out to be of sufficient quality to measure the basic properties of the AGN spectra, such as line strengths and line widths. Therefore, we made use of the opportunity provided by these data to derive statistical information on a well defined complete sample of X-ray selected AGN.
Throughout this paper Seyfert 1 galaxies and QSOs will be discussed jointly since - according to our classification criteria - these two classes are distinguished only by being below or above a certain (arbitrarily chosen) luminosity level. Included in this group are all AGN with permitted emission lines (or components of the permitted emission lines) measurably broader than the forbidden lines and/or showing Fe II emission. Hence Seyfert 1.5-1.9 and NLS1 galaxies (as defined by Véron-Cetty & Véron 2000, following Osterbrock & Pogge 1985) are part of our Seyfert 1 and QSO sample. The (relatively few) Seyfert 2s in the Catalog are discussed separately in Sect. 3.
In order to keep the classification criteria uniform for the whole sample, the [O III ]5007Å /H flux ratio (which is measurable in only part of our spectra) was not used as a classification criterion to discriminate between Seyfert 1s and 2s (Osterbrock & Pogge 1985; Laor 2000). However, in all spectra containing the [O III ]/H region, the corresponding flux ratio was determined and compared to the critical value 3. While all galaxies classified as Seyfert 1s from their line profiles showed (as expected) [O III ]/H, we found (as described in Sect. 3) 6 objects classified as Seyfert 2s with [O III ]/H. Because of this small number the classification of these objects has no influence on the statistical conclusions derived for the Seyfert 1s and QSOs. However, as discussed in Sect. 3, the much smaller Seyfert 2 sample is affected if the [O III ]/H flux ratio is used as classification criterion.
Not discussed in the present paper are AGN number counts, distributions, and the implications of our data for the unresolved X-ray background, since the corresponding results have already been presented by Krautter et al. (1999) and Miyaji et al. (2000). Instead, the present paper will concentrate on conclusions concerning the AGN physics.
© European Southern Observatory (ESO) 2000
Online publication: January 29, 2001