1: \documentstyle[12pt, aaspp4]{article}
2: %\documentstyle[12pt, aasms4]{article}
3: %\documentstyle[referee]{laa} % LaTeX A&A Standard Fonts
4: %\documentstyle{laa} % LaTeX A&A Standard Fonts
5: %\documentstyle{laamt} % LaTeX A&A Monotype Times
6: %\documentstyle{article}
7:
8: %\pagestyle{myheadings}
9:
10: % \topmargin 0cm
11: % \hoffset -2cm
12: % \voffset -2cm
13: % \leftmargin -1.5cm
14: % \textheight 20cm
15: % \textwidth 16cm
16: %\evensidemargin -1cm
17: %\oddsidemargin -0cm
18: \hbadness 10000
19: \vbadness 10000
20: \def\reference{\par\noindent\hangindent 20 pt}
21: \def\ao{$\alpha_{O}$ }
22: \def\hb{ H$_{\beta}$ }
23: \def\ha{H$\alpha$ }
24: \def\arx{$\alpha_{RX} $}
25: \def\sec{^{\prime\prime\!\!}}
26: \def\dvac{de~Vaucouleurs }
27: \def\cosmo{H$_0=50$ km/s/Mpc and q$_0=0\;$ }
28: %\voffset=-1cm
29: %\textheight=11cm
30: \begin{document}
31:
32:
33: \input{psfig.tex}
34:
35: \title{\bf The HST Survey of BL~Lacertae Objects. IV.
36: Infrared Imaging of Host Galaxies}
37:
38: \author{Riccardo Scarpa, C. Megan Urry, Paolo Padovani\footnote{Also affiliated to the
39: Astrophysics Division, Space Science Department, European Space Agency, on
40: leave from Dipartimento di Fisica, II Universit\`a di Roma ``Tor Vergata''.},
41: Daniela Calzetti \& Matthew O'Dowd}
42: \affil{Space Telescope Science Institute}
43: \authoraddr{Space Telescope Science Institute, 3700 San Martin Dr.,
44: Baltimore, MD 21218, USA}
45: \authoremail{scarpa@stsci.edu, cmu@stsci.edu, padovani@stsci.edu,
46: calzetti@stsci.edu, odowd@stsci.edu}
47:
48: \begin{abstract}
49:
50: The HST NICMOS Camera 2 was used for H-band imaging of 12 BL Lacertae objects
51: taken from the larger sample observed with the WFPC2 in the R band
52: (Urry et al. 2000; Scarpa et al. 2000).
53: Ten of the 12 BL Lacs are clearly resolved, and
54: the detected host galaxies are large, bright ellipticals with
55: average absolute magnitude $\langle M_H \rangle =-26.2\pm0.45$~mag and
56: effective radius $\langle r_e \rangle =10\pm 5$~kpc.
57: The rest-frame integrated color of the host galaxies
58: is on average $\langle $R-H$ \rangle =2.3\pm0.3$, consistent with
59: the value for both radio galaxies and normal, non-active elliptical galaxies,
60: and indicating the dominant stellar population is old.
61: The host galaxies tend to be bluer in their outer regions than in their cores,
62: with average color gradient $\Delta (R-H)/\Delta \log r =-0.2$~mag,
63: again consistent with results for normal non-active elliptical galaxies.
64: The infrared Kormendy relation, derived for the first time for BL Lac host
65: galaxies, is $\mu_e = 3.8 \log r_e + 14.8$, fully in agreement with
66: the relation for normal ellipticals.
67: The close similarity between BL Lac host galaxies and normal ellipticals
68: suggests the active nucleus has surprisingly little effect on the host
69: galaxy. This supports a picture in which all elliptical galaxies harbor
70: black holes which can be actively accreting for some fraction of their
71: lifetime.
72:
73: {\underline{\em Subject Headings:}}
74: BL~Lacertae objects --- galaxies: structure --- galaxies: elliptical
75:
76: \end{abstract}
77:
78: \section{Introduction}
79:
80: The properties of BL Lac host galaxies have been the subject of
81: intense study. Early work from the ground
82: found that nearby BL Lacertae objects were surrounded by luminous
83: elliptical galaxies (Ulrich 1989 and reference therein).
84: With higher resolution ground-based observations
85: (Abraham et al. 1991; Stickel et al. 1993; Wurtz, Stocke \& Yee 1996;
86: Kotilainen, Falomo \& Scarpa 1998a; Falomo \& Kotilainen 1999),
87: and now with the 0.1-arcsecond spatial resolution of
88: the Hubble Space Telescope (HST)
89: (Jannuzi, Yanny \& Impey 1997; Yanny, Jannuzi \& Impey 1997;
90: Falomo et al. 1997; Urry et al. 2000; Scarpa et al. 2000),
91: this result has been extended to much higher redshifts.
92: Detection and morphological classification of BL Lac host galaxies
93: is now almost routine for redshifts $z\lesssim 0.5$.
94:
95: Nearly 100 BL Lac host galaxies have been detected so far,
96: and there is now abundant evidence that most, and possibly all,
97: BL Lac host galaxies are ellipticals
98: (Abraham et al. 1991; Wurtz et al. 1996; Scarpa et al. 2000).
99: Although there have been claims of spiral structure
100: in a few host galaxies (Abraham et al. 1991; Wurtz et al. 1996),
101: our HST survey of 110 BL Lac objects
102: found overwhelming preference for $r^{1/4}$-law surface brightness profiles,
103: and no cases where an exponential disk profile was preferred
104: (Falomo et al. 1997; Urry et al. 2000; Scarpa et al. 2000).
105: Given this large sample, the fraction of spiral galaxies among BL Lac hosts
106: must be less than 7\% at the 99\% confidence level (Scarpa et al. 2000).
107: This holds for both radio- and X-ray-selected BL Lac objects, and
108: can be extended (though with less accuracy) to the more powerful
109: flat spectrum radio quasars (Kotilainen, Falomo \& Scarpa 1998b).
110:
111: With high-resolution images in multiple filters, it is possible to
112: probe the stellar populations of AGN host galaxies, and to compare
113: them to normal ellipticals.
114: Indeed, since many normal (non-active) galaxies appear to have
115: central super-massive black holes
116: (Richstone et al. 1998, van der Marel 1999),
117: perhaps all elliptical galaxies have the potential
118: to become active when accretion is turned on.
119: If so, the stellar populations of AGN host galaxies should be
120: the same as those of normal ellipticals.
121:
122: Integrated colors and color profiles have been reported for
123: a handful of BL Lac host galaxies
124: (Urry et al. 1999; Kotilainen, Falomo \& Scarpa 1998a).
125: These sparse data suggest that BL Lac host galaxies
126: are not strongly affected by the intense activity at their centers,
127: a result with obvious implications for the galaxy-AGN connection,
128: and galaxy and AGN evolution.
129:
130: To further investigate the colors of BL Lac host galaxies,
131: we observed 12 BL Lac objects in the H-band with the
132: HST\footnote{Based on observations made with the NASA/ESA
133: Hubble Space Telescope, obtained at the Space Telescope Science
134: Institute, which is operated by the Association of Universities
135: for Research in Astronomy, Inc., under NASA contract NAS~5-26555.}
136: Near-Infrared Camera and Multi-Object Spectrometer (NICMOS).
137: The targets were taken from the larger sample studied previously
138: in the R band with the HST Wide Field and Planetary Camera~2 (WFPC2).
139: For these 12 BL Lac objects we therefore have high-resolution
140: HST observations in two filters probing rather long wavelengths,
141: where the bulk of the emission of the old
142: stellar population lies. These data allow us to derive
143: color profiles and integrated galaxy colors which
144: constrain the stellar content of these host galaxies.
145: The observed sources cover the redshift range $0.1 < z < 0.5$,
146: at the upper end of which cosmic evolution begins to have an effect.
147:
148: In Section 2 we describe the NICMOS observations and data analysis,
149: Section 3 gives the results and discusses them,
150: Section 4 is discussion and conclusions.
151: Throughout the paper we use H$_0=50$ km/s/Mpc and q$_0=0$.
152:
153: \section{Observations and Data Reduction}
154:
155: The HST NICMOS Camera~2 was used to obtain high resolution
156: images of 12 BL Lac objects in snapshot mode (that is, during
157: random gaps in the HST observing schedule). The approved list of
158: targets consisted of 29 BL Lac objects from our original WFPC2
159: snapshot survey list of 132 BL Lacs, comprising seven complete,
160: flux-limited samples selected in the radio, optical, or X-ray.
161: The 29 NICMOS targets had existing WFPC2 F702W data, were
162: uniformly distributed in the redshift range $0.1<z<0.5$, and
163: include both ``red'' (low-frequency-peaked, high-power)
164: and ``blue'' (high-frequency-peaked, low-power) BL Lac jets
165: (Urry 1999).
166: In the end, data were obtained for 8 blue and 4 red BL Lacs.
167: Among them is 0454+844, for which a corrected redshift,
168: $z>1.3$, was subsequently reported.
169:
170: Observations were made in the F160W filter, which has
171: $\lambda_{eff}=1.6 \mu$m and closely matches the standard H-band.
172: Camera~2 has pixel size $0.075$~arcsec,
173: providing at 1.6~$\mu$m Point Spread Function (PSF) sampling as
174: good as the WFPC2 planetary camera (which has pixel size of $0.046$~arcsec)
175: in the R-band.
176: For each BL Lac, we obtained three separate images with increasing
177: exposure times, dithered to improve resolution and to better estimate
178: sky background. The journal of observations is given in Table~1.
179:
180: Data were first reduced and flux calibrated in the standard HST pipeline.
181: The effect of the random bias (known as the ``pedestal'')
182: was then removed as follows:
183: first, all detected sources in the field of view were excluded;
184: then, considering each of four quadrants separately, we found
185: the value of the pedestal that minimized the spread in the remaining
186: (non-source) pixels, with the added constraint that the sky background
187: had to be the same in all four quadrants.
188: Corrected images were then cleaned of cosmic
189: rays and other defects, and a sky frame (obtained from median
190: filtering the three images) was subtracted.
191: To improve the resolution and PSF sampling in the final combined image
192: we then re-sampled each image, splitting each pixel into four sub-pixels.
193: These re-sampled images were re-centered matching the peak of
194: the BL Lac light centroid to within 0.2 pixels, and then combined.
195: The final images are presented as contour plots in Figure~1.
196:
197: The azimuthally averaged, surface brightness radial profiles, shown
198: in Figure~2, were fitted following the same procedure outlined by
199: Scarpa et al. (2000), which is based on a $\chi^2$ minimization strategy.
200: To model the PSF we used the Tiny Tim software (Krist 1995).
201: For the two unresolved sources in our sample (0851+202 and 0454+884),
202: the Tiny Tim PSF model follows the azimuthally averaged
203: surface brightness profile to at least 3.5~arcsec from the nucleus
204: (see Fig.~2), showing it correctly describes the true PSF.
205: In 9 of the other 10 sources, the observed radial profile
206: lies significantly above the PSF model, leaving no doubt that the
207: host galaxy is detected. In the tenth case, 1749+096, the host galaxy
208: is marginally detected; however, the host galaxy was clearly detected
209: in our WFPC2 F702W image (Scarpa et al. 2000) and modeling the
210: excess of light above the Tiny Tim PSF we obtain for the host galaxy
211: a R-H color consistent with the value
212: expected for a normal elliptical galaxy ($R-H=2.3$ at $z=0.32$).
213: Thus we conclude that 1749+096 is resolved and that
214: the Tiny Tim software models the NICMOS Camera~2 PSF very well.
215:
216: \section{Results and Discussion}
217:
218: \subsection{Morphology and Luminosity of the Host Galaxies}
219:
220: Snapshot images are typically shorter than would be ideal
221: for studying extended emission from the host galaxy. Therefore rather than
222: using a full two dimensional approach, we
223: chose to extract azimuthally averaged radial
224: profiles, which can be traced out to $\sim$3--4~arcsec
225: from the center (Fig.~2), significantly increasing the signal-to-noise and
226: providing well-constrained total magnitude and effective radius
227: for the host galaxies, at the expenses of some spatial information like
228: ellipticity and position angle.
229: From the two-dimensional analysis of the R-band data of
230: all $z<0.2$ sources observed in the optical survey, which
231: included 5 of the 12 objects presented here,
232: we know that BL Lac host galaxies are quite round and have undisturbed
233: morphology (Falomo et al. 2000). Thus, we are confident that our one-dimensional
234: approach is not biased or affected by unusual feature of the host galaxy.
235:
236: Ten of the 12 observed BL Lac objects are clearly resolved in the
237: NICMOS images.
238: The two unresolved BL Lacs are the distant source 0454+844 ($z>1.3$),
239: and 0851+202 (OJ~287, $z=0.306$), which has a very luminous nucleus
240: and has so far resisted all attempts to resolve its host galaxy
241: (see Yanny, Jannuzi \& Impey 1997 and counter
242: arguments by Sillanp\"a\"a et al. 1998).
243:
244: The average radial profiles are in all cases well described by a
245: PSF plus a de Vaucouleurs ($r^{1/4}$) law (Figure~2).
246: In no case is a disk galaxy preferred (for 0502+675 and 1749+096
247: the two galaxy models are equally acceptable.)
248: The average extinction and K-corrected absolute H magnitude
249: of the 10 resolved hosts is
250: $\langle M_H \rangle =-26.2\pm0.45$~mag (standard deviation).
251:
252: The average effective radius is $\langle r_e \rangle =10\pm 5$~kpc,
253: and it ranges from 3 to 20~kpc.
254: The derived magnitudes and effective radii are given in Table~2.
255: These results fully confirm what was found from the R-band HST images,
256: that the host galaxies are giant ellipticals,
257: on average $\sim1$~mag brighter than an $M^*$ galaxy
258: ($M_H^*=-25.0\pm 0.3$ galaxy; Mobasher, Sharples \& Ellis 1993).
259:
260: \subsection{Host Galaxy Integrated Color}
261:
262: The extinction-corrected host galaxy total magnitudes, derived from the
263: best fit de Vaucouleurs model allow us to derive integrated
264: R--H colors for all 10 resolved sources (R-band data are in
265: Table~2 of Urry et al. 2000). Results, summarized in Table 3, show
266: that the average BL Lac host galaxy color is R--H=$2.3\pm0.3$ mag,
267: consistent with the value $2.2\pm 0.5$ found for 5 more BL~Lac objects by
268: Kotilainen, Falomo \& Scarpa (1998a), as well as with
269: the value reported for a sample
270: of 12 nearby normal elliptical galaxies, R--H=$2.3\pm 0.2$~mag
271: (Peletier et al. 1990; value converted from V--K assuming
272: V--R=0.58 and H--K=0.22, according to Recilla-Cruz et al. 1990).
273:
274: In the top panel of Figure~3 we plot the R--H
275: colors of our BL Lac host galaxies, computed from extinction and
276: K-corrected magnitudes derived from our WFPC2 and NICMOS data.
277: (In the absence of evolution, the K correction removes the
278: dependence of color on redshift).
279: The BL Lac host galaxy colors cluster around the average color
280: of normal, $z=0$ galaxies (dashed line in Fig.~3).
281:
282: We now compare the BL Lac host galaxies to radio galaxies for which
283: infrared colors are available.
284: From the work of Lilly and Longair (1982),
285: eight 3C radio galaxies (3C33, 98, 123, 192, 219, 295, 299, and 388)
286: have measured R--K colors, $z<0.5$ (as in our BL Lac sample),
287: and are not classified as N-galaxies (whose color is affected by the
288: light from the bright nucleus).
289: After K-correcting the R-band data (the correction is negligible
290: for the K band), and assuming H--K = 0.22 (Recilla-Cruz et al. 1990), we obtain
291: $\langle R-H \rangle =2.3 \pm 0.3$~mag, fully consistent with what we found
292: for BL Lac host galaxies.
293: It is true that the radio galaxies in Lilly \& Longair (1982) are mostly FR~II
294: (high-luminosity) sources, while BL Lac objects are most likely associated
295: with FR~I (low-luminosity) radio galaxies (Urry \& Padovani 1995);
296: however, there is growing evidence that FR~I and FR~II host galaxies
297: have similar optical properties (Govoni et al. 2000; Scarpa \& Urry 2000),
298: so this is unlikely to distort the comparison.
299:
300: Similar colors were reported by De Vries et al. (1998) for a
301: different sample of radio galaxies of various morphological type,
302: at $\langle z \rangle \sim0.3$.
303: They found $\langle R-H \rangle = 2.4$~mag
304: (transformed from observed $\langle R-K \rangle = 3.0$~mag
305: assuming H-K=0.22~mag and K-cor=0.4~mag for the R band; Fukugita et al. 1995),
306: again consistent with the value for BL Lac hosts.
307:
308: The R--H color depends on the age of the dominant stellar
309: population in these galaxies. Due to the metallicity-age degeneracy,
310: the limited information provided by a single color does not allow us
311: to set a tight limit on the epoch of major star formation in these
312: galaxies; however, our data are consistent with the expectation for a
313: single, very old, coeval stellar population. In the bottom panel in
314: Figure~3, the BL Lac host galaxy colors (now not K corrected) are
315: shown to evolve similarly to stellar synthesis models with ages
316: of formation of 4 and 10 Gyrs (solid lines). Similar results have been
317: found for radio galaxies and non-active ellipticals (McCarthy 1993).
318:
319: \subsection{Radial Color Gradients}
320:
321: We derived R--H color profiles for the ten resolved sources
322: combining NICMOS H-band data with the WFPC2 R-band data from
323: Scarpa et al. (2000). Two of the ten (0502+675 and 1749+096)
324: had such noisy profiles, due to
325: the weakness of the host galaxy relative to the central point source,
326: that we do not discuss them further.
327: The eight remaining color profiles are shown in Figure~4.
328: All but one host galaxy show a clear trend for R--H to decrease
329: with increasing radius, that is, the galaxies become bluer in
330: the outer regions.
331: Table~3 gives for each object the slope of the color profile
332: in the (R--H) - Log(r) plane.
333:
334: Similar R--H color gradients for five BL Lac host galaxies were found by
335: Kotilainen, Falomo \& Scarpa et al. (1998a).
336: Urry et al. (1999) reported V--I color profiles for six BL Lac objects
337: based on HST WFPC2 data; because of the smaller wavelength baseline,
338: any color gradients are weaker than in the present R--H color profiles,
339: but the blue-ing trend is still visible in at least
340: three host galaxies (1407+595, 2143-070, and 2254+074).
341:
342: To better quantify this color gradient, we constructed an average
343: R--H color profile, corresponding to the implicit assumption that
344: all host galaxies have the same gradient (a likely oversimplification).
345: We first normalized each profile by subtracting its average color,
346: then averaged the eight normalized profiles, as shown in Figure~5.
347: The error bars represent the standard deviation of the data
348: in each 1-kpc interval, and thus reflect the variation in color from
349: object to object rather than measurement errors.
350: The best-fit linear color gradient has slope
351: $\Delta (R-H)/\Delta \log \ r =-0.2\pm 0.1$~mag,
352: consistent with previous reports
353: ($\Delta (R-H)/\Delta \log \ r =-0.09\pm 0.04$~mag for
354: the five BL Lacs in Kotilainen, Falomo \& Scarpa 1998a).
355:
356: The present data offer strong evidence that BL Lac host galaxies
357: become bluer away from the galaxy center. An immediate consequence
358: (and a consistency check) is that the effective radius of the
359: galaxy is a function of the selected filter.
360: In Figure~6 we compare the best-fit effective radii derived from
361: our HST data, i.e., NICMOS for the H band and WFPC2 for the R band.
362: Even though errors are in some cases large, there is a clear tendency
363: for the effective radius to be larger in the optical than in the infrared.
364: This is clearly because as wavelength decreases a larger fraction of the
365: total light lies in the external regions, i.e., the host galaxies are
366: bluer on their outskirts.
367:
368: Color gradients of the same sign have been found in several studies
369: of normal elliptical galaxies
370: (Boroson et al. 1983; 1987; Cohen 1986; Peletier et al. 1990; Munn 1992).
371: A value of $\Delta (V-K)/\Delta \log \ r = -0.16\pm 0.18$~mag was found by
372: Peletier et al. (1990) for a sample of 12 elliptical galaxies.
373: Given the median redshift of our BL Lac sample ($z=0.2$),
374: our R-band profiles are roughly equivalent to their V-band data at $z=0$;
375: therefore, assuming H and K profiles are similar
376: (since the red stars evolve only slowly),
377: the V--K gradient measured by Peletier et al. is roughly equivalent to
378: our R--H color gradient.
379: The fact that our two values are similar therefore
380: underscores the similarity of BL Lac host galaxies and normal ellipticals.
381: Specifically, the stellar populations of active and non-active
382: ellipticals must not be very different, not only in an integrated sense
383: but differentially across the galaxy.
384:
385: \subsection{The Surface Brightness -- Effective Radius Relation}
386:
387: Infrared observations map the bulk of the slowly evolving stellar populations
388: of galaxies, and are therefore particularly appropriate for investigating
389: luminosity and galaxy morphology. It is therefore useful to investigate
390: the Kormendy relation between effective radius, $r_e$, and
391: surface brightness at that radius, $\mu_e$, in the H band.
392: Plotting our best-fit values for $\mu_e$ and $r_e$ for BL Lac host
393: galaxies (Table~2) shows they follow the same
394: well-defined correlation as non-active elliptical galaxies (Figure~7).
395: The dispersion around the best linear fit is
396: comparable to that found in the R band (Urry et al. 2000).
397: Since the H-band fits are less well constrained (the uncertainties
398: are larger; see Figure~6), the intrinsic correlation must be tighter,
399: reinforcing that infrared observations are indeed
400: optimal for investigating the fundamental plane of ellipticals.
401:
402: A linear fit to our data combined with those from Kotilainen, Falomo \&
403: Scarpa (1998a)
404: gives $\mu_e = 3.8\log r_e +14.8$~mag/arcsec$^2$, where $r_e$ is in
405: kiloparsecs. In the K band, for a sample of 59 normal elliptical galaxies,
406: $\mu_e = 4.3\log r_e +14.3$~mag/arcsec$^2$ has been found (Pahre et al. 1995,
407: after conversion to our adopted cosmology). The agreement is very
408: good considering the restricted range in $r_e$ spanned by BL Lac
409: hosts, which are all giant galaxies. Again, this result fully
410: supports the picture in which all elliptical galaxies
411: have the potential to experience a phase of nuclear activity.
412:
413: \section{Summary and Conclusions}
414:
415: The HST NICMOS Camera~2 was used to image 12 BL Lac objects in the H
416: band. All sources were also observed previously with the HST WFPC2 in
417: the R band. Ten of the 12 BL Lacs were clearly resolved in the H
418: band, with the remaining two having luminous nuclei that probably
419: swamp any surrounding galaxy light. From direct inspection of the
420: NICMOS data (Figure 1), as well as from the R-band WFPC2 data (Falomo
421: et al. 2000), we found no indication of distortion, asymmetry, or
422: tails in the host galaxies, down to surface brightness $\mu_H\sim18.5$
423: mag/asec$^2$. To increase the signal-to-noise ratio and reach fainter
424: structure, we extract for each source the azimuthally averaged radial
425: profile, which is traced out to 3-4 arcsecs from the nucleus down to
426: $\mu_H\sim21$ mag/asec$^2$ (Figure 2). This corresponds to 1-3
427: effective radii depending on galaxy distance and size.
428:
429: The ten detected host galaxies are luminous ellipticals, with
430: $r^{1/4}$-law profiles.
431: Their average K-corrected absolute H magnitude is
432: $\langle M_H \rangle =-26.2$, with $1\sigma$ dispersion of $0.45$~mag,
433: about one magnitude brighter than an M$^*$ galaxy
434: ($M_H^*=-25.0\pm 0.3$ galaxy; Mobasher, Sharples \& Ellis 1993),
435: and their average effective radius is $\langle r_e \rangle =10\pm 5$~kpc.
436:
437: The average R--H color of the BL Lac host galaxies,
438: $\langle R-H \rangle = 2.2 \pm 0.3$~mag,
439: is typical of normal ellipticals or radio galaxies, and indicates
440: the dominant stellar population is more than a few gigayears old.
441: Because of the high resolution of HST, we were able to detect color
442: gradients in the majority of BL Lac host galaxies, such that their
443: cores are redder than their external regions. From our color profiles,
444: which probe a region extending from 0.2 to 1.5 effective radii,
445: we measure an average color
446: gradient of $\Delta (R-H)/\Delta \log \ r =-0.2\pm 0.1$~mag, again
447: similar to those observed in elliptical galaxies (Peletier et al. 1990).
448:
449: The infrared Kormendy relation for BL Lac host galaxies is
450: indistinguishable from the same relation for normal elliptical
451: galaxies. Thus they appear to occupy the same region of the
452: fundamental plane as other ellipticals, including radio galaxies.
453:
454: In every respect we could measure, the host galaxies of BL Lac objects
455: appear to be normal ellipticals. Given the high precision of the HST
456: data, our study was largely free of confusion from the bright nucleus.
457: Had there been dust lanes or other morphological peculiarities beyond
458: $\sim 0.2$~arcsec from the center, they would have been readily
459: apparent. We conclude therefore that to first order, the host galaxy
460: ``knows'' nothing about the active nucleus it harbors. This is in
461: contrast to claims of a nuclear-host connection in more powerful AGN
462: (McLure et al. 1999, McLeod \& Rieke 1995). This result argues
463: against AGN and galaxies being separate entities, and instead supports
464: a picture in which all galaxies can be AGN but with a relatively low
465: duty cycle. This implies all galaxies (at least, elliptical galaxies)
466: have super-massive black holes at their centers, as indeed appears to
467: be the case in many nearby ellipticals (Richstone et al. 1998, van der
468: Marel 1999).
469:
470: Because active galaxies are often observed in clusters, it has been
471: proposed that galaxy mergers are responsible for initiating the
472: nuclear activity. In the HST images of BL Lac objects we do not see
473: any of the morphological peculiarities (e.g., tails, dust lanes) or
474: unusual stellar content which might be expected following a merger. A
475: possibility is that in the vast majority of cases the nuclear activity
476: is triggered by merge of a companion galaxy so small that the global
477: properties of the main galaxy remain unchanged. However, better
478: constraints on this hypothesis would have been possible were we to
479: obtain additional images in a bluer filter. In any case, the present
480: work strongly supports a picture in which all galaxies have the
481: potential to be active. This in turn suggests the formation and
482: evolution of galaxies occurred hand in hand with the growth of
483: super-massive black holes at their centers.
484:
485: \acknowledgements
486: Support for this work was provided by NASA through grant number
487: GO-07893.01-96A from the Space Telescope Science Institute, which is
488: operated by AURA, Inc., under NASA contract NAS~5-26555.
489: This research made use of
490: NASA's Astrophysics Data System Abstract Service (ADS).
491:
492:
493: \noindent {\bf References}
494:
495: \reference Abraham R.G., McHardy I.M. \& Crawford C.S. 1991, MNRAS 252, 482
496: \reference Boroson T.A., Thompson I.B. \& Shectman S.A. 1983, AJ 88, 1707
497: \reference Boroson T.A. \& Thompson I.B. 1987, AJ 93, 33
498: \reference Cohen J.G. 1986, AJ 92, 1039
499: \reference de Vries W.H., O'Dea C.P., Baum S.A. \& Perlman E. 1998, ApJ 503, 156
500: \reference Falomo R. \& Kotilainen J.K. 1999, A\&A 352, 85
501: \reference Falomo R., Urry C.M., Pesce J.E., Scarpa R., Giavalisco M. \&
502: Treves A. 1997, ApJ 476, 113
503: \reference Fukugita M., Shimasaku K. \& Ichikawa T. 1995, PASP 107, 945
504: \reference Govoni F., Falomo R., Fasano G. \& Scarpa R. 2000, A\&A 353, 507
505: \reference Jannuzi B.T., Yanny B. \& Impey C. 1997, ApJ 491, 146
506: \reference Kotilainen J.K., Falomo R. \& Scarpa R. 1998a, A\&A 336, 479
507: \reference Kotilainen J.K., Falomo R. \& Scarpa R. 1998b, A\&A 332, 503
508: \reference Krist, J. 1995, in Astronomical Data Analysis, Software
509: and Systems IV, eds. R. Shaw et al.,
510: (San Francisco: Astr. Soc. Pac.), p. 349
511: \reference Lilly S.J. \& Longair M.S. 1982, MNRAS 199, 1053
512: \reference McCarthy P. 1993, PASP 105, 1051
513: \reference McLeod K.K. \& Rieke G.H. 1995, AJ 109, 1979
514: \reference McLure R.J., Kukula M.J., Dunlop J.S., Baum S.A., O'Dea C.P. \&
515: Hughes D.H. 1999, MNRAS 308, 377
516: \reference Mobasher B., Sharples R. M. \& Ellis R. S. 1993, MNRAS 263, 560
517: \reference Munn J.A. 1992, ApJ 399, 444
518: \reference Pahre M.A., Djorgovski S.G. \& De Carvalho R.R. 1995, ApJL 435, 17
519: \reference Peletier R.F., Valentijn E.A. \& Jameson R.F. 1990, A\&A 233, 62
520: \reference Recillas-Cruz E., Carrasco L., Serrano A. \& Cruz-Gonzales I. 1990,
521: A\&A 229, 64
522: \reference Richstone, D., et al. 1998, Nature 395, 14
523: \reference Scarpa R. \& Urry C. M. 2000, in preparation
524: \reference Scarpa R., Urry C. M., Falomo R., Pesce J. E. \& Treves A. 2000,
525: ApJS, 532, 740
526: \reference Sillanp\"a\"a A., Takalo L.O., Nilsson K., Pursimo T. \&
527: Pietil\"a 1998, in {\it BL Lac Phenomenon}, A.S.P. Conf. Ser.,
528: Vol. 159, ed. L.T. Takalo and A. Sillanp\"a\"a, p. 395
529: \reference Stickel M., Fried J.W., \& K\"uhr H. 1993, A\&AS 98, 393
530: \reference Urry C.M. \& Padovani P. 1995, PASP 107, 803
531: \reference Urry C.M., Scarpa R., O'Dowd M., Falomo R., Pesce J.E.,
532: Treves A. \& Giavalisco M. 1999, ApJ 512, 88
533: \reference Urry C.M., Scarpa R., O'Dowd M., Falomo R., Pesce J.E.,
534: \& Treves A. 2000, ApJ, 532, 861
535: \reference Urry, C.M. 1999 Astroparticle Physics 11, 159
536: \reference Ulrich M.-H. 1989, in ``BL Lac objects', ed. L. Maraschi, T.
537: Maccacaro and M.-H. Ulrich (Springer, Berlin), p. 92
538: \reference van der Marel, R. 1999, AJ 117, 744
539: \reference Wurtz R., Stocke J.T. \& Yee H.K.C. 1996, ApJS 103, 109
540: \reference Yanny B., Jannuzi B.T. \& Impey C. 1997, ApJ 484, L113
541:
542: \newpage
543:
544: {\bf Figure Caption.}
545:
546: {\bf Figure 1:}
547: Contour plots from the central $7.5 \times 7.5$~arcsec regions of the
548: combined NICMOS F160W images of all 12 BL Lac objects.
549: Isophotes are spaced by 0.5 magnitudes and the surface brightness of the fainter
550: is given in each panel (in mag/arcsec$^2$). To improve the visibility of
551: low surface brightness structure, the images
552: have been smoothed using a Gaussian filter with width 0.0375~arcsec.
553: The arrow points north.
554: The host galaxies are resolved in 10 of 12 cases, the two unresolved
555: BL Lacs being 0454+884 and 0851+302. Whatever structure is visible in
556: the latter two images is due to the point spread function, as is
557: the bridge that appears to connect 0607+710 to the nearby star.
558:
559: {\bf Figure 2:}
560: Azimuthally averaged surface brightness radial profile ({\it points with
561: error bars}), and the best-fit point source plus de Vaucouleurs model
562: (or only the point source for unresolved sources).
563: {\it Dotted line:} the point source model (PSF).
564: {\it Dashed line:} de Vaucouleurs model convolved with the PSF.
565: {\it Solid line:} sum of the two model components.
566: When not visible, the dashed line is superimposed on the solid line, and the
567: solid line is covered by the observed points.
568: The PSF model works very well, as evidenced by how it fits the profiles
569: of the unresolved BL Lacs (0454+884 and 0851+202), out to more than
570: 3~arcsec. Nine of the remaining 10 objects have strong excess emission above
571: the PSF; the tenth, 1749+096, has a host galaxy only marginally detected
572: above the PSF but at the level expected given the HST R-band detection.
573:
574: {\bf Figure 3:}
575: BL Lac host galaxy R--H color as a function of redshift,
576: showing their similarity to normal ellipticals with old stellar populations.
577: {\it Solid squares:} our HST data;
578: {\it open squares:} ground-based data from Kotilainen, Falomo \&
579: Scarpa (1998a).
580: {\bf Upper panel:} Rest-frame colors, with both extinction and
581: K corrections applied. The horizontal line at R--H=2.3~mag
582: is the average color observed for normal elliptical galaxies
583: (Kotilainen Falomo \& Scarpa 1998a).
584: {\bf Lower panel:} Observed colors (i.e., extinction- but not K-corrected)
585: of BL Lac host galaxies compared to the expected colors for coeval stellar
586: populations with age 10 Gyrs {\it (upper line)} and 4 Gyrs {\it (lower line)}.
587:
588: {\bf Figure 4:}
589: R-H color profiles of BL Lac host galaxies.
590: Two objects, 0502+675 and 1749+096, are omitted because the
591: signal-to-noise ratios for their host galaxies are very low.
592: With the exception of 0229+200,
593: the host galaxies become significantly bluer in their
594: outer regions, as observed in normal elliptical galaxies
595: (Peletier et al. 1990).
596:
597: {\bf Figure 5:}
598: Average color profile for the 8 well-resolved sources, derived from
599: profiles in Figure~4. Profiles were combined after first subtracting
600: their average color, then computing the weighed average of the points
601: on intervals of 1~kpc. The formal statistical error on each point
602: is smaller than the
603: size of the symbol, while the intrinsic dispersion of the points in
604: each bin is shown at the top-left corner of the figure.
605: The best fit slope {\it (solid line)} is
606: $\Delta (R-H)/\Delta \log \ r =-0.2\pm 0.1$~mag.
607:
608: {\bf Figure 6:}
609: Comparison of effective radii derived from WFPC2 (R-band) and
610: NICMOS (H-band) images. The solid line is the locus of an exact match
611: between the two measures,
612: while the dashed line is a linear fit to the data.
613: This shows that the R-band effective radius is larger than the H-band
614: one, reflecting the fact that the host galaxies become
615: systematically bluer in their outer regions.
616: Two objects show an opposite behavior;
617: for 1749+096, the measurement has large uncertainties,
618: while for 0229+200 the lack of color gradient may be real.
619:
620: {\bf Figure 7:}
621: The H-band $\mu_e - r_e$ relation for BL Lac
622: host galaxies.
623: {\it Solid squares:} our HST data;
624: {\it open squares:} ground-based data from Kotilainen, Falomo \& Scarpa (1998a).
625: {\it Solid line:} linear fit to the data;
626: {\it dashed line:} best-fit relation for a sample of 59
627: normal ellipticals reported by Pahre et al. (1995).
628:
629:
630:
631:
632: \newpage
633:
634: \begin{deluxetable}{llcr}
635: \tablenum{1}
636: \tablewidth{4in}
637: \tablecaption{{\bf Journal of the Observations}}
638: \tablehead{
639: \colhead{BL Lac} & \colhead{Sample/} & \colhead{Date} &
640: \colhead{Exp.Time$^{(a)}$} \\
641: \colhead{Object} & \colhead{Name} & \colhead{Obs.} &\colhead{(sec)}
642: }
643: \startdata
644: 0229+200 & 1ES & 06-Jan-98 & 320 \nl
645: 0454+844 & S5 & 26-Sep-98 & 1920 \nl
646: 0502+675 & 1ES & 04-May-98 & 1428 \nl
647: 0607+710 & MS & 03-Jun-98 & 896 \nl
648: 0706+591 & EXO & 20-Apr-98 & 352 \nl
649: 0851+202 & 1Jy/OJ287& 18-Oct-98 & 896 \nl
650: 1212+078 & 1ES & 10-Mar-98 & 352 \nl
651: 1407+595 & MS & 20-Sep-98 & 1408 \nl
652: 1418+546 & PG & 25-Sep-98 & 352 \nl
653: 1749+096 & PKS & 09-Aug-98 & 1024 \nl
654: 2143+070 & EMSS & 09-Nov-98 & 960 \nl
655: 2356-309 & HEAO-A3 & 14-Oct-98 & 831 \nl
656: \enddata
657: \tablenotetext{(a)}{Total exposure time, i.e., sum of exposure times
658: for all images.}
659: \end{deluxetable}
660:
661: \newpage
662:
663: \begin{center}
664: \scriptsize
665: %\small
666: \hoffset -5cm
667: \begin{deluxetable}{cccccccccccc}
668: \tablenum{2}
669: \tablewidth{6.7in}
670: \tablecaption{{\bf Host Galaxy Properties}}
671: \tablehead{
672: \colhead{BL Lac} & \colhead{z} & \colhead{A$_H^{(a)}$} & \colhead{K Corr.} &\colhead{$m_H^{(b)}$} & \colhead{$m_H^{(c)}$} &
673: \colhead{$m_H^{(d)}$} & \colhead{$\mu_e^{(e)}$} & \colhead{$r_e^{(f)}$} & \colhead{$M_H^{(g)}$} &
674: \colhead{M$_H^{(h)}$} & \colhead{$r_e$} \\
675: \colhead{object} & & & \colhead{H band} &\colhead{(tot)} & \colhead{(PSF)} & \colhead{(host)} &
676: & \colhead{(arcsec)} & \colhead{(PSF)} & \colhead{(host)} & \colhead{(kpc)}
677: }
678: \startdata
679: 0229+200 & 0.139 & 0.13 & 0.01& 13.56 &$ 16.0\pm0.2 $&$ 12.8\pm0.1 $&$ 18.38 $&$ 3.8\pm 0.6$&$ -23.89 $&$ -27.1 $&$ 12 \pm 2 $ \nl
680: 0454+844 &$>$1.34& 0.09 & ... & 16.60 &$ ... $&$ ... $&$ ... $&$ ... $&$ -29.13 $&$ ... $&$ ... $ \nl
681: 0502+675 & 0.314 & 0.16 & 0.03& 18.56 &$ 18.7\pm0.2 $&$ 15.7\pm0.3 $&$ 16.20 $&$ 0.5\pm 0.2$&$ -23.16 $&$ -26.1 $&$ 3.0\pm 1.2 $ \nl
682: 0607+710 & 0.267 & 0.13 & 0.02& 15.05 &$ 16.8\pm0.2 $&$ 14.8\pm0.1 $&$ 18.40 $&$ 1.9\pm 0.3$&$ -24.63 $&$ -26.6 $&$10.5\pm 1.6 $ \nl
683: 0706+591 & 0.125 & 0.08 & 0.01& 13.91 &$ 15.7\pm0.1 $&$ 13.4\pm0.1 $&$ 18.34 $&$ 2.7\pm 0.2$&$ -23.89 $&$ -26.2 $&$ 8.2\pm 0.6 $ \nl
684: 0851+202 & 0.306 & 0.05 & 0.03& 16.60 &$ ... $&$ ... $&$ ... $&$ ... $&$ -25.08 $&$ ... $&$ ... $ \nl
685: 1212+078 & 0.136 & 0.02 & 0.01& 13.92 &$ 15.8\pm0.2 $&$ 13.6\pm0.1 $&$ 17.91 $&$ 2\pm 0.6$&$ -23.93 $&$ -26.1 $&$ 6.5\pm 1.9 $ \nl
686: 1407+595 & 0.495 & 0.02 & 0.11& 16.05 &$ 17.0\pm0.2 $&$ 16.1\pm0.2 $&$ 18.76 $&$ 1.7\pm 0.3$&$ -25.87 $&$ -26.9 $&$13.7\pm 2.4 $ \nl
687: 1418+546 & 0.152 & 0.01 & 0.01& 12.74 &$ 13.0\pm0.1 $&$ 13.7\pm0.2 $&$ 18.84 $&$ 3\pm 0.3$&$ -26.97 $&$ -26.3 $&$10.7\pm 1.1 $ \nl
688: 1749+096 & 0.320 & 0.06 & 0.03& 13.15 &$ 13.2\pm0.1 $&$ 15.3\pm0.3 $&$ 19.98 $&$ 3.3\pm 1.5$&$ -28.60 $&$ -26.5 $&$20 \pm 9 $ \nl
689: 2143+070 & 0.237 & 0.07 & 0.02& 15.22 &$ 16.2\pm0.2 $&$ 15.3\pm0.2 $&$ 18.55 $&$ 1.5\pm 0.4$&$ -24.88 $&$ -25.7 $&$ 7.6\pm 2.0 $ \nl
690: 2356--309& 0.165 & 0.02 & 0.01& 14.41 &$ 15.2\pm0.1 $&$ 14.7\pm0.1 $&$ 18.54 $&$ 1.7\pm 0.2$&$ -24.97 $&$ -25.5 $&$ 6.5\pm 0.8 $ \nl
691: \hline
692: Median & 0.237 & & & & & 14.75 & & & &$ -26.2 $&$ 9.4 $ \nl
693: \enddata
694: \tablenotetext{(a)}{Interstellar extinction in magnitudes, derived assuning $A_H = 0.25 A_R$}
695: \tablenotetext{(b)}{Observed H-band magnitude of the BL Lac source,
696: integrated to the last point of the radial profile shown in Figure~2.
697: Values in this column are neither extinction nor k corrected.
698: Systematic+statistical errors for these values are $\sim 10\% $.}
699: \tablenotetext{(c)}{H-band magnitude of the point source from the best-fit PSF, neither extintion nor k corrected.}
700: \tablenotetext{(d)}{H-band total magnitude of the host galaxy, obtained
701: integrating the best-fit de~Vaucouleurs law to infinite radius. No extinction nor k correction applied.}
702: \tablenotetext{(e)}{H-band surface brightness of the host galaxy at the
703: effective radius, $r_e$, in mag/arcsec$^2$. Values include extinction correction, K-correction,
704: and cosmological dimming.}
705: \tablenotetext{(f)}{Effective radius of the best-fit de~Vaucouleurs law,
706: i.e., radius which encloses half the total light.}
707: \tablenotetext{(g)}{Absolute magnitude of the central point sources, computed for
708: H$_0=50$ km/s/Mpc, q$_0=0$, and corrected for interstellar
709: extinction. K-correction is assumed to be zero because BL Lac nuclei have power law spectra with
710: spectral index $\sim -1$.}
711: \tablenotetext{(h)}{Absolute total magnitudes of the host galaxy, extinction and k-corrected.}
712: \end{deluxetable}
713: \normalsize
714: \hoffset 0cm
715: \end{center}
716:
717:
718: \newpage
719:
720: \begin{deluxetable}{lcc}
721: \tablenum{3}
722: \tablewidth{3.5in}
723: \tablecaption{{\bf Host Galaxy Colors}}
724: \tablehead{
725: \colhead{Object} & \colhead{R--H$^{(a)}$} &
726: \colhead{$\Delta$ (R--H)$/\Delta log(r)$} \\
727: & \colhead{(mag)} & \colhead{(mag)} \\
728: }
729: \startdata
730: 0229+200 &$ 2.5\pm 0.1 $&$ +0.0$ \nl
731: 0502+675 &$ 2.3\pm 0.3 $&$ ...$ \nl
732: 0607+710 &$ 2.3\pm 0.1 $&$ -0.3$ \nl
733: 0706+591 &$ 2.2\pm 0.1 $&$ -0.3$ \nl
734: 1212+078 &$ 2.2\pm 0.1 $&$ -0.6$ \nl
735: 1407+595 &$ 2.1\pm 0.2 $&$ -0.8$ \nl
736: 1418+546 &$ 2.2\pm 0.2 $&$ -0.5$ \nl
737: 1749+096 &$ 2.9\pm 1.3 $&$ ... $ \nl
738: 2143+070 &$ 1.7\pm 0.2 $&$ -0.6$ \nl
739: 2356-309 &$ 2.2\pm 0.1 $&$ -0.3$ \nl
740: \hline
741: Mean Profile &$ $&$ -0.2$ \nl
742: \enddata
743: \tablenotetext{(a)}{Integrated rest frame R--H color, K-corrected and
744: corrected for extinction. Color gradients are expressed in
745: magnitudes per decade, i.e., it is the variation of color
746: for $\Delta log(r)=1$.}
747: \end{deluxetable}
748:
749:
750:
751: \newpage
752:
753: \begin{figure}
754: \psfig{file=figure1a.ps,width=14cm}
755: \caption{
756: }
757: \end{figure}
758:
759: \begin{figure}
760: \psfig{file=figure1b.ps,width=14cm}
761: \end{figure}
762:
763: \begin{figure}
764: \vspace{-1cm}
765: \psfig{file=figure2a.ps,width=14cm}
766: \caption{
767: }
768: \end{figure}
769:
770: \begin{figure}
771: \psfig{file=figure2b.ps,width=14cm}
772: \end{figure}
773:
774: \begin{figure}
775: \psfig{file=figure3.ps,width=11cm}
776: \caption{
777: }
778: \end{figure}
779:
780: \begin{figure}
781: \psfig{file=figure4.ps,width=14cm}
782: \caption{
783: }
784: \end{figure}
785:
786: \begin{figure}
787: \psfig{file=figure5.ps,width=14cm}
788: \caption{
789: }
790: \end{figure}
791:
792:
793: \begin{figure}
794: \psfig{file=figure6.ps,width=13cm}
795: \caption{
796: }
797: \end{figure}
798:
799: \begin{figure}
800: \psfig{file=figure7.ps,width=14cm}
801: \caption{
802: }
803: \end{figure}
804:
805: \end{document}
806:
807: