1: \documentclass[manuscript]{aastex}
2: %\documentclass[preprint]{aastex}
3: %\usepackage{emulateapj5}
4: \usepackage[]{natbib}
5: %\usepackage{apjfonts}
6: \usepackage{graphics}
7: \usepackage{amssymb}
8: \usepackage{amsmath}
9:
10: \slugcomment{} \shorttitle{0846+51W1} \shortauthors{ZHOU ET AL.}
11:
12: \begin{document}
13:
14: \title{Hybrid nature of 0846+51W1: a BL Lac object with a narrow line Seyfert 1 nucleus}
15: \author{Hong-Yan Zhou, Ting-Gui Wang, Xiao-Bo Dong, Cheng Li, and Xue-Guang Zhang }
16: \email{mtzhou@ustc.edu.cn}
17:
18: \affil{Center for Astrophysics, University of Science and Technology of
19: China, Hefei, 230026, China}
20:
21:
22: \begin{abstract}
23: We found a NLS1 nucleus in the extensively studied eruptive BL
24: Lac, 0846+51W1, out of a large sample of NLS1 compiled from the
25: spectroscopic dataset of SDSS DR1. Its optical spectrum can be
26: well decomposed into three components, a power law component from
27: the relativistic jet, a stellar component from the host galaxy,
28: and a component from a typical NLS1 nucleus. The emission line
29: properties of 0846+51W1, $FWHM(H\beta)\simeq 1710~ km~ s^{-1} $
30: and $\frac{[OIII]\lambda 5007}{H\beta}\simeq 0.32$ according to
31: its SDSS spectrum observed when it was in faint state, fulfil the
32: conventional definition of NLS1. Strong FeII emission is detected
33: in the SDSS spectrum, which is also typical of NLS1s. We try to
34: estimate its central black hole mass using various techniques and
35: find that 0846+51W1 is very likely emitting at $a~few \times 10\%$
36: of Eddington luminosity. We speculate that Seyfert-like nuclei,
37: including NLS1s, might be concealed in a significant fraction of
38: BL Lacs but have not been sufficiently explored due to the fact
39: that, by definition, the optical-UV continuum of such kind of
40: objects are often overwhelmed by the synchrotron emission.
41: \end{abstract}
42:
43: \keywords{galaxies: active --- galaxies:Seyfert --- galaxies:BL
44: Lacertae --- quasars:indevidual (0846+51W1) --- radiation: lines,
45: continuum}
46:
47: \section{INTRODUCTION}
48: In 1985, Osterbrock \& Pogge identified a special class of active
49: galactic nuclei (AGN) denominated as Narrow Line Seyfert 1
50: galaxies (NLS1s), which are characterized by their narrow Balmer
51: emission line width of $FWHM(H\beta)\lesssim 2000~km~s^{-1}$ and
52: forbidden to permitted line ratio $\frac{[OIII]\lambda
53: 5007}{H\beta}\lesssim \frac{1}{3}$ (for the conventional
54: definition of NLS1s, see Pogge 2000). Succedent studies reveal
55: their extreme properties, such as the narrowest $H\beta$ line
56: width, the weakest $[OIII]\lambda 5007$, the strongest FeII
57: emission and the steepest soft X-ray slope. These characteristics
58: locate NLS1s at the extreme end of the Eigenvector 1 (E1)
59: parameter space (Boroson \& Green 1992; Sulentic et al. 2000) and
60: their unusualness is very useful to test the viabilities of AGN
61: models.
62:
63: It is found that radio-loud AGN occupy a restrict E1 parameter
64: space opposite to NLS1s which are usually radio-quiet (Sulentic et
65: al. 2003). Only three radio-loud NLS1s with radio-loudness
66: (defined as $\frac{f_{radio}}{f_{optical}}$) between $\sim 10-100$
67: are known before 2000. They are all radio-intermediate sources
68: according to the criterion of Sulentic et al (2003). About twenty
69: more radio-loud NLS1s were identified by Whalen et al. (2001) in
70: the FIRST Bright Quasar Survey. Zhou \& Wang (2002) found eight
71: more radio-loud NLS1s by cross-correlating the Veron \&
72: Veron-Cetty AGN with the FIRST and NVSS radio catalog. Again,
73: almost all of these newly dig out radio-loud NLS1s are moderate
74: radio sources. It is remarkable that all of the discovered
75: radio-loud NLS1s are compact at the present spatial resolution,
76: indicating that relativistic beaming might be important in at
77: least some of these objects. SDSS J0948+0022 is hitherto
78: the only genuinely very radio-loud NLS1 possibly with a
79: relativistic jet beaming toward the observer (Zhou et al. 2003).
80: These properties call to remembrance of blazar, which is another
81: small divertive subset of AGN.
82:
83: Blazar is a pictorial term first proposed by Spiegel in 1978 be
84: applied to rapidly variable objects (see Burbidge \& Hewitt 1992)
85: and all of the known blazars are radio sources. Nowadays it is
86: believed that blazars, including OVVs (optically violent
87: variables) and BL Lacs, are those AGN that have a strong
88: relativistically beamed component close to the line of sight. BL
89: Lacs and OVVs share many common properties except that, by
90: definition, in the former emission line is very weak or absent.
91: However, the rarity of very radio-loud NLS1s, whose occurrence in
92: low redshift broad line AGN is estimated $\lesssim 0.2\% $ (Zhou
93: et al. 2003) is in the way addressing ourselves to this problem.
94:
95: The large sky area coverage and moderate deepness of the Sloan
96: Digital Sky Survey (SDSS\footnote{The SDSS Web site is
97: http://www.sdss.org/.}, York et al. 2000) make it be propitious to
98: exploring rare objects such as very radio-loud NLS1s. In this
99: letter we report the discovery of another such object,
100: 0846+51W1=SDSS J084957.98+510829.1, out of $\sim 500$ NLS1s
101: compiled from the spectroscopic dataset of the SDSS Data Release 1
102: (DR1). It is even more conspicuous than SDSS J094857.3+002225 and
103: shows many dramatic properties.
104:
105: Actually 0846+51W1 was originally found by Arp et al. (1979) and
106: has been the subject of much study since then. This object is
107: violently variable in its optical flux ($\Delta V \sim 5^{m}$ over
108: a time span of $\sim 1$ year and $\Delta V \sim 4^{m}$, from
109: $V\simeq 15^{m}.8$ to $V\sim 19^{m}.5$ within one month). At its
110: maximum light burst, the optical spectrum was found to be
111: featureless, while emission lines were detected when it became
112: fainter. Its optical slope may vary dramatically from
113: $\alpha_{o}\approx 1.6$ when the object is bright to
114: $\alpha_{o}\approx 2.8$ when it is faint ($f_{\nu}\varpropto
115: \nu^{-\alpha_{o}}$, c.f. Arp et al. 1979; Stickel et al. 1989). It
116: is also found to be highly polarized in both of the radio and
117: optical bandpass (Moore \& Stockman 1981; Sitko et al. 1984).
118: Therefore 0846+51W1 bears all the characteristics of BL Lac
119: object. However, the narrow wavelength coverage of Arp et al. and
120: Stickel et al. led 0846+51W1 be taken as a high redshift BL Lacs
121: (z=1.86) by these authors whereas the SDSS spectrum clearly shows
122: that its true redshift is $z=0.5835$.
123:
124: We will analysis the SDSS optical spectrum of 0846+51W1 in detail
125: in \S 2. Some implication of 0846+51W1 is discussed in \S 3. The
126: main purpose of this letter is to reignite further interest to
127: this fascinating object. We adopt a $\Lambda$-dominated cosmology
128: with $H_0=70~km~ s^{-1}~Mpc^{-1}$, $\Omega_M=$ 0.3 and
129: $\Omega_\Lambda=$ 0.7 through out this letter.
130:
131:
132: \section{OPTICAL SPECTRUM ANALYSIS}
133:
134: 0846+51W1 is observed by SDSS only because it is the counterpart
135: of a FIRST radio source and its SDSS spectrum is shown in Figure 1
136: (denoted as thin line) with recognizable emission lines labelled.
137: At first sight, the optical spectrum of 0846+51W1 looks like a
138: typical NLS1 and by no means let any association with a BL
139: Lacertae occur to us though the spectrum is rather noisy. Almost
140: all of the familiar emission lines, even the weak ones such as
141: [NeV]$\lambda 3346$, [NeV]$\lambda 3426$, [NeIII]$\lambda 3869$
142: and [NeIII]$\lambda 3888$ are present in the spectrum. Optical and
143: UV FeII multiplets can also be easily spotted out. The spectrum
144: were taken at 2002 November 29 and the visual magnitude estimated
145: from the spectrum is about $19^{m}.8$ indicating that 0846+51W1
146: was in its quiescent state when it was observed. At a redshift of
147: $z=0.5835$, its luminosity is $M_{V}\simeq -22^{m}.1$, which is
148: comparable to luminous CD galaxies indicating that the host may
149: contribute significantly to the observed spectrum. Many authors
150: invoke gravitational (micro)lenses to interpret the eruptive
151: behavior of 0846+51W1 (e.g., Nottale 1986; Stickel et al. 1989).
152: However, the HST image when 0846+51W1 was in its faint state
153: ($V\simeq 19^{m}.7$) did not show significant resolved structure
154: (Bahcall et al. 1993) and, with broader wave coverage and higher
155: resolution, the SDSS spectrum does not show the signature of any
156: intervening galaxy. Because the S/N ratio of the SDSS spectrum is
157: not high, we model it with three components
158: \begin{equation}\label{eq1}
159: f(\lambda)=aA(\lambda)+bB(\lambda)+cC(\lambda)
160: \end{equation}
161: where $A(\lambda)$ denotes a composite spectrum of NLS1 (cf.
162: Constantin \& Shields 2003), $B(\lambda)\varpropto \lambda^{-0.5}$
163: is a power law component from the relativistic jets, $C(\lambda)$
164: is the template of elliptical galaxy (cf. Mannucci et al. 2001)
165: and a, b, and c represent the relative contribution of the three
166: components which are set free. Emission lines except FeII
167: multiplets are masked in the fitting procedure. The final fit is
168: done through minimization of $\chi^{2}$ and the result is
169: acceptable (reduced $\chi^{2}=1.36$) and is also shown in Figure
170: 1.
171:
172: Now the continuum subtracted spectrum is fitted to measure
173: prominent emission line parameters. The $H\beta +[OIII]\lambda
174: \lambda 4959,5007$ regime is fitted with one Lorentzian + three
175: Gaussians. The $H\beta $ is modelled by one Lorentzian (broad
176: component) and one Gaussian (narrow component) and the
177: $[OIII]\lambda \lambda 4959,5007 $ doublet are fitted with two
178: Gaussians. The width and redshift of the three Gaussians are
179: forced to be the same and the intensity ratio of the
180: $[OIII]\lambda \lambda 4959,5007 $ doublet is fixed to the
181: theoretical value. We also force the ratio of $[OIII]\lambda 5007$
182: to $H\beta $ narrow component to be equal to 10, which is the
183: typical value of Seyferts. This is a common procedure to prevent
184: the fitting routine to yield non-physical value when spectrum is
185: noisy (e.g., Veron et al. 2001). Almost all known NLS1s show
186: strong FeII emission as does 0846+51W1. However, measurement of
187: its FeII multiplets is rather challenged for the present spectral
188: quality. Judging from the fitting residual of Equation (1), what
189: we can say is that its optical FeII strength should be at least
190: comparable to, possibly stronger than that of typical NLS1s. We
191: did not analyses other emission lines because the quality of the
192: spectrum is not high enough for us to draw any significant
193: conclusion. The main parameters of prominent emission lines are
194: listed in Table 1.
195:
196:
197: \begin{table*}
198: \begin{minipage}{180mm}
199: {\bf Table 1.} Emission line parameters of 0846+51W1. \\[0.1cm]
200: \nobreak
201: \begin{tabular}{lccc}
202: \hline \hline
203: line & Flux & FWHM \\
204: & 10$^{-17}$erg s$^{-1}$ cm$^{-2}$ & $km s^{-1}$ \\
205: \hline
206: $[OIII]\lambda 5007$ & 45$\pm $4.7 & 372$\pm $31 \\
207: $H\beta $(broad component) & 139$\pm $22 & 1710$\pm $184 \\
208: $MgII\lambda $2800 & 286$\pm $39 & 2512$\pm$471 \\
209: FeII$\lambda \lambda $4570 & $\gtrsim$ 100 & \\
210:
211: \hline
212: \end{tabular}
213: \end{minipage}
214: \end{table*}
215:
216: \section{DISCUSSION}
217:
218: \subsection{THE CENTRAL BLACK HOLE MASS AND THE ACCRETION RATE OF 0846+51W1}
219: Mass estimates for the central black holes in AGN have become
220: feasible using various techniques and the exact value of $M_{BH}$
221: of 0846+51W1 is of much interesting considering its dualism of
222: NLS1 and BL Lac object. We will try to estimate $M_{BH}$ using
223: three distinct approaches and make comparison among the results
224: yielded.
225:
226: We first use the virial assumption with FWHM(H$\beta$) measured
227: and the decomposed luminosity of the disk component ("A" component
228: of \S2). Then $M_{BH}$ can be estimated as (Kaspi et al. 2000)
229: \begin{equation}\label{eq2}
230: M_{BH}=1.464\times10^{5}(\frac{R_{BLR}}{lt-days})(\frac{v_{FWHM(H\beta)}}{10^{3}km
231: s^{-1}})^{2}~ M_{\odot}
232: \end{equation}
233: where $R_{BLR}$ is the size of broad $H\beta$ emission line region
234: which is related to the monochromatic luminosity of "A" component
235: through the empirical relation (Kaspi et al. 2000 converted to our
236: adopted cosmology)
237: \begin{equation}\label{eq3}
238: R_{BLR}=22.3[{\frac{\lambda L_{\lambda}(5100\AA)}{10^{44}erg
239: s^{-1}}}]^{0.7}~ lt-days
240: \end{equation}
241: For the measured value of $S_{\lambda}(5100\AA)\simeq 1.2\times
242: 10^{-17}~ erg~ s~ cm^{-2}~\AA $ ("A" component) and
243: $FWHM(H\beta)\simeq 1710~km~ s^{-1}$ we obtain $M_{BH}\simeq
244: 8.2\times 10^{6}~M_{\odot}$.
245:
246: The second method to estimate $M_{BH}$ is to use $[OIII]\lambda
247: 5007$ as a surrogate for the stellar velocity dispersion of the
248: bulge $\sigma_{*}$ and the $M_{BH}-\sigma_{*}$ correlation
249: (Tremaine et al. 2002)
250: \begin{equation}\label{eq4}
251: log(\frac{M_{BH}}{M_{\odot}})=(8.13\pm0.06)+(4.02\pm0.32)log(\frac{\sigma_{*}}{200~km~s^{-1}})
252: \end{equation}
253: If the motion of [OIII] emission line clouds in the NLR of AGN is
254: dominated by the gravitational potential of the host galaxy bulge,
255: a strong correlation between FWHM[OIII] and $\sigma_{*}$ would be
256: expected and such a correlation is indeed found by Nelson \&
257: Whittle (1996). Nelson (2000) suggested that FWHM[OIII] may be
258: used as a surrogate for $\sigma_{*}$ by the relation
259: \begin{equation}\label{eq5}
260: \sigma_{*}\sim \frac{FWHM[OIII]}{2.35}
261: \end{equation}
262: Using Equation (3) and Equation (4) we obtain $M_{BH}\simeq
263: 5.2\times 10^{7}~ M_{\odot}$.
264:
265: The third way to estimate $M_{BH}$ is to use the
266: $M_{BH}$-$M_{bulge}$ correlation. The $M_{bulge}$ can be deduced
267: adopting the $L/M$ ratio of elliptical galaxies where L correspond
268: to the "C" component of \S2. Using the the following empirical
269: relation (Laor 2001 converted to our adopted cosmology)
270: \begin{equation}\label{eq6}
271: M_{V}(bulge)=-10.06\pm 1.08 - (1.38\pm 0.13)
272: log(\frac{M_{BH}}{M_{\odot}})
273: \end{equation}
274: with the decomposed host luminosity of $M_{V}(bulge)\simeq
275: -20^{m}.5$, we got $M_{BH}\simeq 4.3\times 10^{7}~ M_{\odot}$.
276:
277: Considering the large uncertainty of these techniques ($\sim
278: 0.4-0.7$ dex, Vestergaard 2004), the estimated values of $M_{BH}$
279: are consistent with each other. Because its nonthermal emission is
280: highly boosted, it is difficult to estimate the bolometric
281: luminosity of 0846+51W1 and we use the emission of "A" component
282: to yield its lower limit. Assuming that the bolometric luminosity
283: is about nine times of the monochrome luminosity at B-band (Elvis
284: et al. 1994), the Eddington mass $M_{Edd}$ should be $> 5.3\times
285: 10^6 M_{\odot}$. Hence 0846+51W1 should emit at $>10\% $ of
286: Eddington luminosity.
287:
288:
289: \subsection{On the hybrid nature of 0846+51W1}
290: The SDSS spectrum of 0846+51W1 is typical of NLS1s. The width of
291: $H\beta $, $FWHM(H\beta)\approx 1700 km s^{-1}$ is narrower than
292: normal broad line AGN and the flux ratio of $[OIII]/H\beta \approx
293: 0.3$ excludes the possibility that it might be a type 2 AGN. The
294: primary ionization source should be the thermal component from the
295: accretion disk because the little blue bump clearly presents
296: itself. However, its soft X-ray photon index $\Gamma_{0.2-2.4
297: keV}=0.61^{+0.38}_{-0.49}$ is very flat. Such a difference between
298: 0846+51W1 and "normal" NLS1s can be anticipated because in the
299: X-ray we may be observing mainly the jet emission as in other BL
300: Lacs.
301:
302: Considering that NLS1s show extreme characteristics opposite to
303: classical radio-loud AGN, the nature of radio sources in 0846+51W1
304: is of particular interesting. Accumulated evidence indicates that
305: the physical driver of Eigenvector 1 may be the accretion rate of
306: active nuclei, with source orientation playing a concomitant role
307: and NLS1s are believed to have large $\dot{m}\equiv
308: \frac{\dot{M}}{M_{BH}}$ and small inclination angle $i$. It may be
309: exactly the case of 0846+51W1 and the other very radio-loud NLS1,
310: SDSS J0948+0022. The radio power of 0846+51W1 $P_{5 GHz}\sim
311: 3.9\times 10^{26} W~ Hz^{-1}$ (radio fluxes are adopted from Arp
312: et al. 1979 and Gregory \& Condon 1991) is comparable to SDSS
313: J0948+0022, which is also a highly variable object (Zhou et al.
314: 2003). It has been found that in quasars with no Doppler boosting,
315: the luminosity of $H\beta $ is tightly correlated with the
316: continuum luminosity with the median rest-frame $H\beta $
317: equivalent width $EW_{H\beta} \sim 80 $ \AA~ (Veron-Cetty \& Veron
318: 2000). We note that $EW_{H\beta}\simeq 18 $ \AA~ of 0846+51W1 and
319: 25 \AA~ of SDSS J0948+0022 are both much less than the above
320: median value. This indicates that synchrotron emission from
321: relativistic jet might make significant contribution to the
322: optical continuum. Indeed, we would have $EW_{H\beta}\simeq 73$
323: \AA for 0846+51W1 if the underling continuum of "A" component is
324: used in calculation. We argue that 0846+51W1 and SDSS J0948+0022
325: are actually NLS1s oriented with jet axis almost along our line of
326: sight and consequently extremely beamed and henceforth show
327: blazar-like behavior. If the Doppler factor is $\gtrsim 10$, the
328: intrinsic radio luminosity of these two objects would be around
329: the FR I/FR II transition.
330:
331: Arp et al. (1979) found that 0846+51W1 fulfils four of the five
332: criteria for BL Lac object, i.e., weak line feature, large
333: amplitude variability, nonthermal continuum and red color, except
334: that polarization measures were not available then and strongly
335: argued that 0846+51W1 should be classified as a BL Lac object.
336: High polarization of $> 10\% $ was also detected in the optical
337: and radio by Moore \& Stockman (1981) and Sitko et al. (1984).
338: According to the current unified schemes of AGN, the parent
339: population of blazar are radio galaxies (RGs) and radio-loud
340: quasars (RLQs). The radio morphology of RGs and RLQs are
341: classified to two categories according to Fanaroff \& Riley
342: (1974): FR IIs are the classical double radio sources with
343: edge-bright lobes while FR Is have edge-darkened morphologies. BL
344: Lacs are taken as "beamed" FR I radio galaxies while OVVs as
345: beamed FR IIs. The main deference between the two is believed to
346: be the accretion rate, which is rather small in the former. For
347: $\dot{m}\lesssim 1\%$, the accretion flow is advection dominated
348: with small radiative efficiency. In this scenario, neither FR I
349: nor BL Lac can be associated with optically powerful quasar.
350: However, broad line AGN and high luminous quasar associated with
351: FR I radio structure have been reported in recent year (Lara et
352: al. 1999; Blundell \& Rawlings, 2001). Correspondingly, broad
353: emission lines have also been detected in dozens of BL Lacs and
354: even in BL Lacertae itself (see Table 3 and 4 of Veron-Cetty \&
355: Veron 2000).
356:
357: BL Lacs are conventionally defined as blazars with rest-frame
358: emission line equivalent widths smaller than 5 \AA~ (Morris et al.
359: 1991). The distribution of emission line equivalent width in
360: blazar is obviously not bimodal and the strength of the highly
361: variable synchrotron continuum can further blur such an arbitrary
362: boundary. Objects that fulfil the above criterion may form a
363: rather complex family. While the emission line properties of some
364: BL Lacs mimic LINERs indicating low mass accretion rate, other BL
365: Lacs including the prototype one may harbor a Seyfert-like nucleus
366: (e.g., Corbett et al. 2000). Some of the BL Lacs with Seyfert-like
367: nucleus may have relatively high accretion rate but, in optical-UV
368: band, the emission lines and thermal continuum from the hot
369: accretion disc can be overwhelmed by the Doppler-boosted
370: synchrotron component some of the time. 0846+51W1 is the rare care
371: at it happens when the synchrotron continuum became faint enough
372: for the NLS1 nucleus to reveal itself. In accordance with
373: 0846+51W1 which is likely emitting at near Eddington luminosity,
374: Blundell \& Rawlings (2001) found that a optical powerful quasar
375: E1821+643 is associated with a FR I radio structure indicating
376: that relatively high accretion rate can also occur in FR Is.
377:
378: It is remarkable that, on the one hand, no highly radio-luminous
379: FR Is have been found as yet, and on the other hand, no NLS1 has
380: been reported to be very powerful in the radio except SDSS
381: J0948+0022 and 0846+51W1 which are strongly boosted and their
382: intrinsic radio luminosity may not be too high. This suggests that
383: apart from other properties such as the spin rate of the central
384: black hole, mass accretion rate play an role in determining the
385: radio power and morphology. When $\dot{m}\ll \dot{m}_{Edd}$, only
386: low power jet can engender because there is not enough input
387: energy, which can be easily disrupted and dissipated within short
388: distance from the core and forms an FR I source (De Young 1993).
389: Contrarily, under the condition of near or supper Eddington
390: accretion, neither does the air rich environment lend itself to
391: the collimation and propagation of the jet, which may be
392: translated into outflow in some extreme cases. In 0846+51W1 and
393: SDSS J0948+0022 we are very likely observing the innermost part of
394: the jet pointing toward us and the fact that all of the radio
395: sources in NLS1s are compact can be understood according to this
396: interpretation. We speculate that, while both low and high mass
397: accretion can occur in FR Is and their beamed cousins BL Lacs, the
398: accretion rate of FR IIs can only (at one time) be moderately
399: high. It has been pointed by Blundell \& Rawlings (2001) that
400: optically luminous FR I quasars are well under-investigated. We
401: speculate that Seyfert-like nuclei might be concealed in a
402: significant fraction of BL Lacs but have not been sufficiently
403: explored due to the fact that, by definition, the optical-UV
404: continuum of such kind of objects are often overwhelmed by the
405: synchrotron emission. A few of these objects may be "very"
406: radio-loud NLS1s provided that they are observed at very small
407: inclination angle.
408:
409:
410:
411: \acknowledgments This work was supported by Chinese NSF through
412: NSF10233030 and NSF19990754, and a key program of Chinese Science
413: and Technology ministry. This paper has made use of the data from
414: the SDSS. Funding for the creation and the distribution of the
415: SDSS Archive has been provided by the Alfred P. Sloan Foundation,
416: the Participating Institutions, the National Aeronautics and Space
417: Administration, the National Science Foundation, the U.S.
418: Department of Energy, the Japanese Monbukagakusho, and the Max
419: Planck Society. The SDSS is managed by the Astrophysical Research
420: Consortium (ARC) for the Participating Institutions. The
421: Participating Institutions are The University of Chicago,
422: Fermilab, the Institute for Advanced Study, the Japan
423: Participation Group, The Johns Hopkins University, Los Alamos
424: National Laboratory, the Max-Planck-Institute for Astronomy
425: (MPIA), the Max-Planck-Institute for Astrophysics (MPA), New
426: Mexico State University, Princeton University, the United States
427: Naval Observatory, and the University of Washington.
428:
429:
430: \begin{thebibliography}{}
431: \bibitem[Arp, Sargent, Willis, \& Oosterbaan(1979)]{1979ApJ...230...68A} Arp, H., Sargent, W.~L.~W., Willis, A.~G., \& Oosterbaan, C.~E.\ 1979, \apj, 230, 68
432: \bibitem[Blundell \& Rawlings(2001)]{2001ApJ...562L...5B} Blundell, K.~M.~\& Rawlings, S.\ 2001, \apjl, 562, L5
433: \bibitem[Boroson \& Green(1992)]{1992ApJS...80..109B} Boroson, T.~A.~\& Green, R.~F.\ 1992, \apjs, 80, 109
434: \bibitem[Burbidge \& Hewitt(1992)]{1992vob..conf....4B} Burbidge, G.~\& Hewitt, A.\ 1992, Variability of Blazars, 4
435: \bibitem[Constantin \& Shields(2003)]{2003PASP..115..592C} Constantin, A.~\& Shields, J.~C.\ 2003, \pasp, 115, 592
436: \bibitem[Corbett, Robinson, Axon, \& Hough(2000)]{2000MNRAS.311..485C} Corbett, E.~A., Robinson, A., Axon, D.~J., \& Hough, J.~H.\ 2000, \mnras, 311, 485
437: \bibitem[Fanaroff \& Riley(1974)]{1974MNRAS.167P..31F} Fanaroff, B.~L.~\& Riley, J.~M.\ 1974, \mnras, 167, 31P
438: \bibitem[Gregory \& Condon(1991)]{1991ApJS...75.1011G} Gregory, P.~C.~\& Condon, J.~J.\ 1991, \apjs, 75, 1011
439: \bibitem[Kaspi et al.(2000)]{2000ApJ...533..631K} Kaspi, S., Smith, P.~S., Netzer, H., Maoz, D., Jannuzi, B.~T., \& Giveon, U.\ 2000, \apj, 533, 631
440: \bibitem[Laor(2001)]{2001ApJ...553..677L} Laor, A.\ 2001, \apj, 553, 677
441: \bibitem[Lara et al.(1999)]{1999NewAR..43..643L} Lara, L., M{\' a}rquez, I., Cotton, W.~D., Feretti, L., Giovannini, G., Marcaide, J.~M., \& Venturi, T.\ 1999, New Astronomy Review, 43, 643
442: \bibitem[Mannucci et al.(2001)]{2001MNRAS.326..745M} Mannucci, F., Basile, F., Poggianti, B.~M., Cimatti, A., Daddi, E., Pozzetti, L., \& Vanzi, L.\ 2001, \mnras, 326, 745
443: \bibitem[Moore \& Stockman(1981)]{1981ApJ...243...60M} Moore, R.~L.~\& Stockman, H.~S.\ 1981, \apj, 243, 60
444: \bibitem[Morris et al.(1991)]{1991ApJ...380...49M} Morris, S.~L., Stocke, J.~T., Gioia, I.~M., Schild, R.~E., Wolter, A., Maccacaro, T., \& della Ceca, R.\ 1991, \apj, 380, 49
445: \bibitem[Nelson \& Whittle(1996)]{1996ApJ...465...96N} Nelson, C.~H.~\& Whittle, M.\ 1996, \apj, 465, 96
446: \bibitem[Nelson(2000)]{2000ApJ...544L..91N} Nelson, C.~H.\ 2000, \apjl, 544, L91
447: \bibitem[Nottale(1986)]{1986A&A...157..383N} Nottale, L.\ 1986, \aap, 157, 383
448: \bibitem[Osterbrock \& Pogge(1985)]{1985ApJ...297..166O} Osterbrock, D.~E.~\& Pogge, R.~W.\ 1985, \apj, 297, 166
449: \bibitem[Pogge(2000)]{2000NewAR..44..381P} Pogge, R.~W.\ 2000, New Astronomy Review, 44, 381
450: \bibitem[Sitko, Rudnick, Jones, \& Schmidt(1984)]{1984PASP...96..402S} Sitko, M.~L., Rudnick, L., Jones, T.~W., \& Schmidt, G.~D.\ 1984, \pasp, 96, 402
451: \bibitem[Stickel, Fried, \& Kuehr(1989)]{1989A&A...224L..27S} Stickel, M., Fried, J.~W., \& Kuehr, H.\ 1989, \aap, 224, L27
452: \bibitem[Sulentic et al.(2003)]{2003ApJ...597L..17S} Sulentic, J.~W., Zamfir, S., Marziani, P., Bachev, R., Calvani, M., \& Dultzin-Hacyan, D.\ 2003, \apjl, 597, L17
453: \bibitem[Sulentic, Zwitter, Marziani, \& Dultzin-Hacyan(2000)]{2000ApJ...536L...5S} Sulentic, J.~W., Zwitter, T., Marziani, P., \& Dultzin-Hacyan, D.\ 2000, \apjl, 536, L5
454: \bibitem[V{\' e}ron-Cetty \& V{\' e}ron(2000)]{2000A&ARv..10...81V} V{\' e}ron-Cetty, M.~P.~\& V{\' e}ron, P.\ 2000, \aapr, 10, 81
455: \bibitem[Vestergaard (2004)]{} Vestergaard M.\ 2004, astro-ph/0401436
456: \bibitem[Whalen, Laurent-Muehleisen, Moran, \& Becker(2001)]{2001AAS...199.5001W} Whalen, J., Laurent-Muehleisen, S.~A., Moran, E.~C., \& Becker, R.~H.\ 2001, Bulletin of the American Astronomical Society, 33, 1373
457: \bibitem[York et al.(2000)]{2000AJ....120.1579Y} York, D.~G.~et al.\ 2000, \aj, 120, 1579
458: \bibitem[Zhou \& Wang(2002)]{2002ChJAA...2..501Z} Zhou, H.~\& Wang, T.\ 2002, Chinese Journal of Astronony and Astrophysics, 2, 501
459: \bibitem[Zhou et al.(2003)]{2003ApJ...584..147Z} Zhou, H., Wang, T., Dong, X., Zhou, Y., \& Li, C.\ 2003, \apj, 584, 147
460:
461: \end{thebibliography}
462:
463: \clearpage
464:
465: \begin{figure}
466: \epsscale{1.0} \plotone{f1.eps} \caption{Optical continuum of
467: 0846+51W1 (the uppermost thin line) can be well decomposed into
468: three components of A: average NLS1 spectrum, B: spectral index of
469: $\alpha =1.5$ power law spectrum ($f_{\nu}\varpropto
470: \nu^{-\alpha}$) and C: elliptical template. The uppermost thick
471: line is the sum of the three components.} \label{reduction1}
472: \end{figure}
473:
474:
475: \begin{figure}
476: \epsscale{1.0} \plotone{f2.eps} \caption{Both of $H\beta $ and
477: MgII can be well fitted with Lorentz profile and the line widths
478: are typical of NLS1s.} \label{reduction2}
479: \end{figure}
480:
481:
482:
483: \end{document}
484: