1 Introduction

According to the commonly accepted paradigm, Active Galactic Nuclei (AGN) are thought to harbour supermassive black holes (BH) surrounded by geometrically thin accretion disks, the latter triggering the formation of relativistic jets. Increasing evidence now indicates, however, that this picture probably has to be expanded: first, hierarchical galaxy evolution schemes suggest, that as a result of mergers between galaxies, binary BH systems (BBHSs) should be generally expected in the center of elliptical galaxies, and hence, for example, in the typical hosts of BL Lac objects (cf. Begelman et al. 1980; Kauffmann 1996; Richstone et al. 1998; Milosavljevic & Merritt 2001; Yu 2002). Secondly, from a phenomenological point of view, a multitude of observational evidence has been plausibly related to the presence of BBHSs in AGN, including the observed misalignment, precession and wiggling of extragalactic jets (e.g. Conway & Wrobel 1995; Kaastra & Roos 1992), periodic optical outburst events as observed in the quasar OJ 287 (e.g. Sillanpää et al. 1988; Lehto & Valtonen 1996; Villata et al. 1998) and the helical motion of knots (e.g. Abraham & Carrara 1998; Britzen et al. 2001).

One expects that a binary framework for AGN may be particularly relevant for our understanding of BL Lac objects such as Mkn 501 (e.g. De Paolis et al. 2002).

Mkn 501 (z=0.034) is one of at least four AGN which have been reliably detected at TeV energies (see Catanese & Weekes 1999 for a review). From optical observations its host is known to be the elliptical galaxy UGC 10599 (cf. Stickel et al. 1993). As a BL Lac object, Mkn 501 belongs to the blazar class of AGN which are thought to have relativistic jets oriented at a small angle to the line of sight, so that the observed flux is strongly enhanced by relativistic beaming effects. Detailed observations since 1997 have confirmed the picture of powerful activity in Mkn 501 and revealed a wide range of variability patterns during outbursts (e.g. Protheroe et al. 1998) including flaring episodes of several days and rapid variability on subhour timescale (e.g. Sambruna et al. 2000). The evidences for a 23-day periodicity in both the TeV and the X-ray lightcurves during the 1997 high state (cf. Hayashida et al. 1998; Kranich et al. 1999; Nishikawa et al. 1999) may be counted among the most fascinating features (see Kranich 2001 and Kranich et al. 2001 for a recent assessment of significance) and possibly indicates the presence of a BBHS in Mkn 501 (cf. Rieger & Mannheim 2000). In this paper we analyse the implications of the apparent disagreement of independent central mass estimates for Mkn 501. Starting with an investigation of the constraints given by host galaxy observations (Sect. 2), we proceed with an analysis of the robustness of the HE emission estimates (Sect. 3) and finally consider the consequences expected from a binary BH modelling (Sect. 4).