BL Lacs

The BL Lac objects found within DXRBS fall mostly in the range $10^{-7.5} \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}L_X/L_R \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}10^{-5.5}$. Only a few objects are found at higher values of L_X/L_R, as expected given our radio flux limits (§ 2.1). A large fraction of the BL Lacs so far in our sample are at L_X/L_R < 10^-6.5, the region of Figure 6 populated by LBLs. In this region of parameter space (to the left of the left-most dashed line in Figure 6), DXRBS includes objects up to two orders of magnitude fainter than the 1 Jy survey. Most of these objects are radio galaxies (which are discussed in more detail in §7), but three are clearly BL Lacs. As with the FSRQs, this is expected given our much fainter flux limits. This comment can also be made for objects in the range $10^{-6.5} \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}L_X/L_R \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}10^{-5.5}$, between the two dashed lines in Figure 6 - of which all but one object is classified herein as a BL Lac.

About 50% of the BL Lacs so far in our sample fall in the range $10^{-6.5} \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}L_X/L_R \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}10^{-5.5}$ ($0.72 \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}\alpha_{\rm rx} \lower.5ex\hbox{$\; \buildrel < \over \sim \;$}0.85$), and are ``intermediate'' BL Lacs, objects with spectral shapes intermediate between the X-ray bright and radio-bright varieties of BL Lacs (Padovani & Giommi 1995a). Similar objects have also been found in two ROSAT All-Sky Survey-based samples (Kock et al. 1996, Nass et al. 1996), as well as the Einstein Slew Survey sample (Perlman et al. 1996a). For comparison, we have plotted the $(\alpha_{\rm ox},\alpha_{\rm ro})$ values for members of these three ROSAT-based surveys (DXRBS, HQS/RASS and RC) in Figure 7. Since redshift information for the ROSAT-based samples is still incomplete, it is difficult to compare them on the (L_X,L_R) plane. It is important to note that the optical magnitudes used to derive the effective spectral indices at present include the galaxy contribution; this explains the somewhat extreme objects in the lower right corner of the diagram, all BL Lacs and radio galaxies at relatively low redshifts. If only the non-thermal flux were used, these objects would move towards the other objects along lines parallel to the dashed lines in Fig. 7.

As can be seen from Figure 7, each of these surveys covers a slightly different region of the $(\alpha_{\rm ox},\alpha_{\rm ro})$ plane. The HQS/RASS survey of Nass et al. (1996) is dominated by objects at $\alpha_{\rm rx} < 0.72$ (the left-most diagonal line plotted in Figure 7), i.e. HBLs, but does include a significant fraction of these intermediate objects (8 of 34). Its makeup is thus similar to the Slew Survey sample (Perlman et al. 1996a), which contains 5 transition objects and 5 LBLs ($\alpha_{\rm rx} \gt 0.85$, the right-most line plotted in Figure 7) among a sample of 66 (Figure 6; note that the diagonal lines thereon plotted represent the same values of $\alpha_{{\rm rx}}$). By comparison, objects from the RC survey of Kock et al. (1996) are more heavily concentrated (6 of 13) at intermediate values of $\alpha_{{\rm rx}}$,although another 6 are HBLs. The fraction of intermediate BL Lacs among the DXRBS sample (15 of 32) is comparable to that in the RC sample. However, as shown in Figure 7, the DXRBS intermediate BL Lacs are concentrated at lower values of $\alpha_{\rm ro}$ than those in either the HQS/RASS or RC samples. The large majority of the remainder (12 of 32) of the DXRBS BL Lacs are LBLs, and only a few objects (5 of 32) are HBLs.

These differences are no doubt due to the differing flux limits of the surveys. The similarities of the HQS/RASS and Slew Survey samples are no surprise given their low radio flux limits (a few mJy) and coverage of mostly X-ray bright objects. By comparison, the RC sample covers a range of X-ray fluxes similar to the HQS/RASS sample, but does not go as deep as DXRBS (by a factor $\sim 10$), while its radio flux limit, at $\sim 40$ mJy, is similar to ours. Finally, both the HQS/RASS and RC survey groups observed only objects with optical counterparts on sky survey plates, a restriction not found in DXRBS. An examination of Figure 7 reveals that these facts naturally translate to the $(\alpha_{\rm ox},\alpha_{\rm ro})$ plane.

What is most important in Figures 6 and 7 is that once again the advantages of newer, deeper surveys which cover large dynamic ranges of fluxes in more than one survey band is shown. Only these very recent surveys (and particularly DXRBS, which already contains more intermediate BL Lacs than the HQS/RASS and RC samples combined) have revealed a large population of BL Lacs with $0.72 < \alpha_{\rm rx} < 0.85$; they went largely undetected in the 1 Jy and EMSS surveys because of the single-band nature and small dynamic range covered by those surveys (Stickel et al. 1994, Stocke et al. 1991, 1997). The exact population fraction of these ``intermediate'' BL Lacs is not yet known, as a bivariate luminosity function has yet to be computed for the BL Lac class. Our results do not allow us to comment significantly on the relative proportion of HBLs and LBLs among BL Lacs (e.g., Padovani & Giommi 1995a; Urry & Padovani 1995), since we are sensitive to high L_X/L_R objects only at high X-ray fluxes.