Gd071.tex
1: %                                                                 aa.dem
2: % AA vers. 5.2, LaTeX class for Astronomy & Astrophysics
3: % demonstration file
4: %                                                 (c) Springer-Verlag HD
5: %                                                revised by EDP Sciences
6: %-----------------------------------------------------------------------
7: %
8: %\documentclass[referee]{aa} % for a referee version
9: %
10: %\documentclass[onecolumn]{aa}
11: \documentclass{aa}
12: \usepackage{graphicx}
13: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
14: \usepackage{txfonts}
15: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
16: %
17: \begin{document}
18: %
19:    \title{Nuclear star formation in the quasar PG1126$-$041 from adaptive
20:    optics assisted spectroscopy \thanks{Based on data obtained
21:    at the VLT through the ESO program 71.B-0453(A).}}
22: 
23: 
24:    \author{G. Cresci \inst{1} \and R. Maiolino \inst{2} \and A. Marconi \inst{2}  
25:           \and F. Mannucci \inst{3} \and G.~L. Granato \inst{4}}
26: 
27:    \offprints{G. Cresci}
28: 
29:    \institute{Dip. di Astronomia - Universit\`a di Firenze,
30:    	Largo E. Fermi 5, I-50125 Firenze, Italy;
31:               \email{gcresci@arcetri.astro.it}
32:          \and INAF, Osservatorio Astrofisico di Arcetri, 
33:            Largo E. Fermi 5, I-50125, Firenze, Italy
34: 	 \and Istituto di Radioastronomia, sezione di Firenze,
35:            Largo E. Fermi 5, I-50125, Firenze, Italy
36: 	 \and INAF, Osservatorio Astronomico di Padova, 
37: 	 Vicolo dell'Osservatorio 5, I-35100 Padova, Italy}
38: 
39:    \date{Received ; accepted }
40: 
41: 
42:    \abstract{We present adaptive optics assisted spectroscopy of three quasars
43:    obtained with NACO at VLT. The high angular resolution achieved with the
44:    adaptive optics ($\sim0.08''$), joined to the diagnostic power of near-IR
45:    spectroscopy,
46:    allow us to investigate the properties of the innermost 100~pc of
47:    these quasars.
48:    In the quasar with the best adaptive optics correction, PG1126$-$041, 
49:    we spatially resolve the Pa$\alpha$ emission within the
50:    nuclear 100~pc. The comparison with higher excitation lines suggests that
51:    the narrow Pa$\alpha$ emission is due to nuclear star formation.
52:    The inferred intensity of the nuclear star formation
53:    (13~$\rm M_{\odot}~yr^{-1}$) 
54:    may account for most of the far-IR luminosity observed in this quasar.
55: 
56:       \keywords{Galaxies: active -- Galaxies: nuclei -- quasars: emission lines --
57:       		Galaxies: starburst -- Techniques: high angular resolution --
58:                 Instrumentation: adaptive optics}
59:    }
60: 
61:    \maketitle
62: %
63: %________________________________________________________________
64: 
65: \section{Introduction}
66: 
67: The study of the stellar populations and star formation
68: in the host galaxies of Active Galactic Nuclei (AGNs) is 
69: fundamental to understand the connection between black-hole
70: growth and galaxy formation.
71: Several evidences for such a connection have been obtained in
72: local low-luminosity AGNs, i.e. Seyfert galaxies
73: (e.g. Cid Fernandes et al. \cite{cid01}, Heckman et al. \cite{heckman97},
74: Maiolino et al. \cite{maiolino95}, Oliva et al. \cite{oliva99}).
75: At higher, quasar-like luminosities the
76: investigation of star formation and stellar population is much
77: more difficult because the tracers of star formation are heavily
78: diluted by the strong active nucleus. Nevertheless a few observational programs
79: have been successful in disentangling the host galaxy
80: and star formation from the quasar light. The separation
81: of the two components has been achieved either through a
82: spectroscopic decomposition or through high angular resolution
83: imaging observations (e.g.
84: Canalizo \& Stockton \cite{canalizo01},
85: Courbin et al. \cite{courbin02}, Dunlop et al. \cite{dunlop03},
86: Jahnke et al. \cite{jahnke03}, Schade et al. \cite{schade00}).
87: By using such observating strategies,
88: evidence both for active star formation and relatively
89: quiescent hosts was found. However, all these studies probed 
90: the stellar population on scales of a few kpc, while the stellar
91: activity within the central 100~pc, where any interplay with the quasar
92: activity is expected to occur, remains poorly explored.
93: 
94: During recent years the performances of adaptive optics systems have improved
95: significantly,
96: achieving very high angular resolutions ($\la 0.1''$) even on relatively 
97: faint sources. 
98: Such high angular resolution, coupled with spectroscopic capabilities
99: of some of these systems, allows us to efficiently disentangle components
100: associated with stellar activity even in the circumnuclear region of
101: luminous quasars. In this letter we present NACO observations of
102: three quasars at z $\sim$0.06 selected to have an excess of far-Infrared
103: emission. In one of the quasars we find evidence for star formation
104: within the central 100 pc with a rate of $\rm \sim 13~M_{\odot}~yr^{-1}$, 
105: which may account for most of its far-IR emission.
106: We assume a "standard" cosmology, with $\rm H_0 = 70~km~s^{-1}~Mpc^{-1}$,
107: $\rm \Omega _m =0.3$ and $\rm \Omega _{\Lambda}=0.7$.
108: 
109: %__________________________________________________________________
110: 
111: \section{Sample selection, observations and data reduction}
112: 
113: The targets were selected from the sample
114: of Andreani et al. (\cite{andreani01}) who provide mid-IR and far-IR data for a large number
115: of optically selected quasars.
116: We selected quasars matching the following constraints:
117: 1) z$\sim$0.06, so that the CONICA projected resolution
118: on the source is better than 100~pc (in the K band) and, at the
119: same time, both Pa$\alpha$ and [SiVI]1.97$\mu$m emission lines
120: are shifted into the K band; 
121: 2) far-IR luminosity $\rm L_{FIR}>10^{44}~erg~s^{-1}$, i.e.
122: objects more IR luminous than the average of quasars at the same
123: redshift and, therefore, suspected to harbor starburst activity;
124: 3) radio quiet, to avoid synchrotron contamination to the far-IR
125: radiation; 
126: 4) visible magnitude V$<$14.5 mag, so that the Adaptive Optics correction can
127: achieve good performances with Strehl ratios of $\sim$0.2 
128: by using the quasar itself as a reference to close the loop.
129: Four quasars were selected and three of them, whose
130: properties are listed in Tab.~1, were observed in service mode with NACO.
131: %
132: \begin{table*}
133: 	\label{osservazioni}
134: 	\begin{center}
135: 	\begin{tabular}{l c c c c c c c c c}
136: 	\hline
137: 	\hline
138: 	Name & RA(J2000) & DEC(J2000) & z & m$_V$ & L$_{\rm FIR}$ & Date & seeing & FWHM & SR \\
139: 	\hline
140: 	IRAS09149$-$6206 & 09 16 09.4 & $-$62 19 29 & 0.057 & 14.1 & 7.9 & Apr 20, 2003 & 1.0$''$ & 0.09$''$ (100 pc) & 5\% \\
141: 	PG1126$-$041 & 11 29 16.6 & $-$04 24 08 & 0.060 & 14.5 & 3.2 & Jun 17, 2003 & 0.4$''$ & 0.07$''$ (81 pc) & 26\% \\
142: 	PG2214+139 & 22 17 12.3 & +14 14 21 & 0.066 & 14.5 &  1.5 & Jun 18, 2003 & 0.8$''$ & 0.09$''$ (114 pc) & 11\% \\
143: %	PG1211+14 & 12 14 17.7 & +14 03 12 & 0.080 & 14.2 &  & 0.5'' &  & \\
144: 	\hline
145: 	\hline
146: 	\end{tabular}
147: 	\caption{List of the bserved quasars, along with their properties and log of
148: 	the observations. L$_{\rm FIR}$ is in units of
149: 	$\rm 10^{44}erg~s^{-1}$;
150: 	\textit{seeing} is the median DIMM seeing 
151: 	during the observation;
152: 	\textit{FWHM} is the Full Width Half Maximum of the PSF on the Adaptive Optics corrected images, in arcsec 
153: 	and in parsec; \textit{SR} is the Strehl Ratio measured on the images.}
154: 	\end{center}
155: \end{table*}
156: %
157: Both images and spectra were obtained in the K band. Images were
158: obtained in the IB\_2.18 filter (to avoid saturation)
159: with the S27 camera, yielding a pixel scale of 
160: $0.027''$/pix. Spectra were obtained with a $0.086''$ slit
161: at PA$=$0$^{\circ}$, with the Grism2 
162: coupled with the SK order sorting filter,
163: yielding a spectral resolution $\rm R=\lambda/\Delta \lambda=1400$,
164: and with the S54 camera yielding a pixel scale along the slit of $0.054''$/pix.
165: During each observing run we obtained a diffraction limited nuclear
166: spike  of $\sim0.08''$ (corresponding to $\sim$100~pc projected on these
167: quasars). In the spectroscopic mode the pixel size
168: undersamples the PSF. The Strehl Ratio ranges from 5\% to 26\% (see Tab.~1).
169: 
170: The spectra were reduced following the standard steps for NIR spectroscopy. The spectra obtained 
171: at different positions along the slit were subtracted from each other to remove the background, 
172: then flat fielded, aligned, co-added and calibrated.  
173: Atmospheric features were corrected by dividing for the spectrum of a reference star. 
174: B3III and B5V stars were used for PG1126 and IRAS09149 respectively, and
175: in these cases
176: the spectra were then multiplied by a blackbody at 20000~K to
177: re-establish the correct slope of the 
178: continuum. For PG2214 we used a G0V star, and the spectra were then
179: multiplied by the 
180: solar spectrum to remove the stellar features and continuum slope
181: (see Maiolino et al. \cite{maiolino96} for details). Images were instead
182: reduced using the \textit{eclipse} software and the CONICA pipeline.
183: 
184: \section{Analysis and results}
185: 
186: In this letter we focus on the analysis of the spectra, since they provide the most
187: interesting information.
188: Of the three quasar spectra we obtain spatially resolved features only for the
189: case 
190: with the best Adaptive Optics correction, i.e. PG1126$-$041 which was observed with a
191: Strehl Ratio of 26\%. The integrated spectrum is shown in Fig.~\ref{spectra}. In the 
192: following we will concentrate on the analysis of this spectrum. 
193: 
194: %
195: \begin{figure}[!h]
196: 	\centering
197: 	\resizebox{\hsize}{!}{\includegraphics{Gd071_f1.eps}}
198: 	\caption{Continuum subtracted spectra of PG1126$-$04. The Br$\gamma$* 
199: 	introduced by the division of the reference B star is marked.}
200:         \label{spectra}
201: \end{figure}
202: %
203: Pa$\alpha$ is clearly broad and certainly dominated by the Broad Line Region (BLR).
204: It is more difficult to indentify the presence of a narrow component associated
205: either with a Narrow Line Region (NLR) or with star formation.
206: A method to investigate the presence of a narrow component,
207: while exploiting the high angular resolution delivered by the AO,
208: is to study the spatial variation of the Pa$\alpha$ profile along
209: the slit. The BLR is certainly unresolved at our resolution (R$<$1pc) and
210: therefore the profile of the Pa$\alpha$, if totally attributed to the BLR,
211: should not change spatially along the slit, and should simply scale
212: in intensity following the profile of the PSF. We have investigated the
213: variation of the Pa$\alpha$ profile along the slit by
214: subtracting the continuum of all spectra at different locations
215: along the slit and then rescaled the spectra to match the Pa$\alpha$
216: flux on the central spectrum. In Fig.~\ref{compare} we show this comparison for
217: the spectrum $0.054''$ north of the central spectrum for PG1126: the
218: solid line shows the (continuum subtracted) central spectrum, while
219: the dashed line shows the rescaled off-nuclear spectrum. The lower
220: panel shows the difference between the two spectra which, although noisy,
221: shows the presence of a excess narrow component ($\rm FWHM \sim 200\ km/s$)
222: in the off-nuclear spectrum. This demonstrates the presence of
223: a narrow component of Pa$\alpha$,
224: associated either with a NLR or with star forming regions. 
225: 
226: Once proved the existence of a narrow component, separated
227: from the BLR, we can follow its spatial behavior along with
228: other spectral features.
229: We have fitted the Pa$\alpha$ of PG1126 profile with two
230: broad gaussians, which reproduce the broad (BLR) component, and a narrow
231: component. The resulting fit for the central spectrum is shown in Fig.~\ref{fit}. 
232: %
233: \begin{table}
234: 	\label{parametri}
235: 	\begin{center}
236: 	\begin{tabular}{l c c c c}
237: 	\hline
238: 	\hline
239: 	Line & Flux & v$-$v$_{\rm sys}$ & FWHM  \\
240: 	\hline
241: 	\textbf{Pa$\alpha$} - Broad 1 & 953.3  $\pm$ 43.3   & 14 $\pm$ 37  & 3003 $\pm$ 134 \\
242: 	\textbf{Pa$\alpha$} - Broad 2 & 457.1  $\pm$ 41.7   & 37 $\pm$ 32  & 1312 $\pm$ 142 \\
243: 	\textbf{Pa$\alpha$} - Narrow & 27.6   $\pm$ 7.0    & 71 $\pm$ 32  & Unresolved \\ %183  $\pm$ 212 \\
244: 	\hline
245: 	\textbf{Br$\delta$} - Broad  & 129.0  $\pm$ 10.4  & $-$75 $\pm$ 191 & 2277 $\pm$ 258 \\
246: 	\textbf{Br$\delta$} - Narrow & 13.8   $\pm$ 6.3   & $-$124 $\pm$ 123  & 697  $\pm$ 413 \\
247: 	\hline
248: 	\textbf{[SiVI]} - Broad  & 89.7   $\pm$ 6.8   & $-$203 $\pm$ 333 & 2797 $\pm$ 240 \\
249: 	\textbf{[SiVI]}	- Narrow & 8.0    $\pm$ 1.7   & 42 $\pm$ 89  & 360  $\pm$ 185 \\
250:   	\hline
251: 	\hline
252: 	\end{tabular}
253: 	\caption{Best fit parameters for the components of the central spectrum of PG1126$-$041,
254: 	for Pa$\alpha$, 
255: 	Br$\delta$ and [SiVI]. \textit{Flux} is in units of $10^{-16}$ erg s$^{-1}$ cm$^{-2}$; 
256: 	the shift the various component from the systemic velocity
257: 	(\textit{v$-$v$_{sys}$}) and their \textit{FWHM} are in km s$^{-1}$.}
258: 	\end{center}
259: \end{table}
260: %
261: \begin{figure}%[!ht]
262: 	\centering
263: 	\resizebox{\hsize}{!}{\includegraphics{Gd071_f2.eps}}
264: 	\caption{Comparison between the central spectrum and the spectrum $0.054''$
265: 	to the north, around 
266: 	the Pa$\alpha$ region.
267: 	In the \textit{upper panel} the solid gray line shows the (continuum subtracted) 
268: 	central spectrum, while the black dashed line shows the rescaled off-nuclear spectrum. 
269: 	The \textit{loer panel} shows the difference between the two spectra.}
270:         \label{compare}
271: \end{figure}
272: %
273: In Tab.~2 we list the best fit parameters for the components. 
274: For what concerns the blend of [SiVI]1.96$\mu$m and Br$\delta$, each of the two lines
275: was fitted by using two gaussian components only (additional components are not
276: required given the lower S/N for these fainter lines)\footnote{Note that the profile
277: of the Pa$\alpha$ cannot be used to fit the Br$\delta$ because different hydrogen
278: lines originating from the BLR have often different profiles, like in this case.}. 
279: [SiVI] in particular is
280: well fitted by a broad and a narrow component. The presence of both narrow and broad
281: components of coronal lines (such as [SiVI]) is well known in AGNs. A component
282: with a width intermediate between BLR and NLR ($\rm \sim 1000-2000\ km~s^{-1}$)
283: is commonly observed and ascribed to an ``Intermediate Emission Line Region''
284: (Giannuzzo et al. \cite{giannuzzo95}). The narrow component of the coronal lines
285: comes instead
286: from the ``classical'' NLR and extends for a few 100~pc
287: (Maiolino et al. \cite{maiolino00}, Thompson et al. \cite{thompson01}).
288: The broad component requires also the profile to be asymmetric,
289: and specifically with a blue wing.
290: Such a blue asymmetry is also commonly observed
291: on high excitation lines from the nuclear region of several Seyferts
292: (e.g. Marconi et al. \cite{marconi96}, Oliva et al. \cite{oliva94}).
293: %
294: \begin{figure}
295: 	\centering
296: 	\includegraphics[width=0.5\textwidth]{Gd071_f3.ps}
297: 	\caption{Fit of the Pa$\alpha$ profile for the central spectrum of PG112.
298: 	The fit is overplotted 
299: 	on the original spectrum, together with the single gaussian components (see text).}
300:         \label{fit}
301: \end{figure}
302: % 
303: 
304: The line profiles on the off-nuclear spectra were fitted by keeping frozen
305: the parameters of the broad components (except for a scaling factor)
306: while the narrow components were left free to vary. In Fig.~\ref{andamento} we show
307: the intensity of the various lines as a function of the offset from the nucleus.  
308: The broad component of the Pa$\alpha$ gives the profiles of the
309: PSF along the slit (and it is also consistent with the profile of the
310: continuum, not shown). The most interesting result, expected from the
311: analysis discussed above, is that the intensity
312: of the narrow component of Pa$\alpha$ does not follow the PSF profile and shows an
313: excess of emission to the North, tracing an emission line region extending
314: for 0.05$''$--0.1$''$, or 60--120~pc, from the nucleus.
315: 
316: We investigate the nature of the narrow Pa$\alpha$ emitting region by 
317: comparing its flux with the observed [SiVI] emission. The narrow component
318: of the latter traces the NLR as it is not produced by star forming regions, 
319: while the narrow Pa$\alpha$ may either
320: come from star formation or from the NLR. The different trends of
321: the narrow Pa$\alpha$ and [SiVI] in Fig.~\ref{andamento} suggest that the two lines
322: have different origins. We have further investigated this issue by
323: comparing the ratio between narrow Pa$\alpha$ and [SiVI] in the
324: narrow line region of Sy2 galaxies observed by other authors
325: (Sy1 cannot be used because the hydrogen lines include the broad
326: component and generally a decomposition of narrow and broad components
327: is not given). Pa$\alpha$ is not observable
328: in low redshift Sy2s, but it can be easily derived from Br$\gamma$
329: by using the
330: case B recombination ratio (which holds for the NLR):
331: Br$\gamma$/Pa$\alpha$=0.082. The additional problem
332: is however that in Sy2 galaxies Br$\gamma$ may well be contributed by star
333: formation, since there are several evidences for nuclear and circumnuclear
334: star formation, as discussed in Sect.~1. For this reason we decided to focus
335: on two specific templates: NGC~1068 and Circinus. These are very
336: nearby Sy2's, extremely well studied, for which there is no evidence in the
337: central region for significant contribution to Br$\gamma$ by active star
338: formation (star formation occurs in rings at a radius of $\sim10''-30''$).
339: The Pa$\alpha$/[SiVI] ratio inferred for the NLR is 2.2 in Circinus
340: (Maiolino et al. \cite{maiolino98}) and 1.1 in NGC~1068
341: (Reunanen et al. \cite{reunanen03}).
342: In PG1126 the ratio between narrow Pa$\alpha$/[SiVI] is 3.4 on the
343: nucleus and 6.4 in the northern region at 60~pc from the nucleus. 
344: This comparison indicates that there is a strong excess of narrow Pa$\alpha$ 
345: emission with respect to that expected from a NLR.
346: Such a Pa$\alpha$ excess is probably 
347: due to active star formation in the nuclear and circumnuclear region, within
348: the central $\sim$100~pc.
349: %
350: \begin{figure}%[!h]
351: 	\centering
352: 	\resizebox{\hsize}{!}{\includegraphics{Gd071_f4.ps}}
353: 	\caption{Intensity of the various lines as a function of the offset from the nucleus, 
354: 	normalized at the flux in the central spectrum. The peak emission of the narrow Pa$\alpha$ 
355: 	is not aligned with the nucleus, but shows an excess $\sim$60 pc to the north.}
356:         \label{andamento}
357: \end{figure}
358: %
359: 
360: Assuming that all the narrow Pa$\alpha$ is associated with star
361: formation we can estimate the star formation rate in the central
362: region of this quasar. Unfortunately, our data do not provide a
363: bi-dimensional information of the Pa$\alpha$ distribution
364: in the central region. However, we can roughly estimate the circumnuclear
365: integrated Pa$\alpha$ emission by assuming that the emission is well represented by the 
366: average of the northern and southern sides in our spectrum. The inferred total
367: narrow Pa$\alpha$ flux estimated in this way is
368: $\rm F_{tot}(Pa\alpha _{Narrow}) = 2.15~10^{-14}~erg~s^{-1}cm^{-2}$.
369: From case B recombination ($\rm Pa\alpha/H\alpha=0.107$), 
370: we infer a total H$\alpha$ luminosity due to star formation of
371: $\rm L(H\alpha)_{SF}=1.7~10^{42}~erg~s^{-1}$. By using the relation
372: between H$\alpha$ luminosity and star formation rate given in
373: Kennicutt et al. (\cite{kennicutt98}) we derive a nuclear star formation rate of
374: $\rm SFR \approx 13~M_{\odot}$ $\rm yr^{-1}$. According to the relation
375: between SFR and far-IR emission obtained by Kennicutt et al. (\cite{kennicutt98}),
376: such a star formation rate is expected to produce a far-IR
377: luminosity of $\rm \sim 3~10^{44}~erg~s^{-1}$, which matches the luminosity 
378: observed from this quasar (Tab.~1). Therefore, the nuclear star formation
379: detected in the nuclear few 100~pc may well account for most 
380: of the far-IR luminosity emitted by this quasar and could
381: explain the excess of far-IR emission in this object (Sect.~2). 
382: A similar result was found at higher redshift by Alexander et al. (\cite{alexander}),  
383: who derived that star formation appears to dominate the bolometric output 
384: of AGNs hosted in bright SCUBA galaxies.
385: 
386: Of course, the estimated nuclear star formation rate is derived from a
387: monodimensional spectrum and, in particular, the inferred nuclear (R$<$250~pc)
388:  Pa$\alpha$ luminosity has required a large aperture correction
389: (a factor of $\sim$3). As consequence, a confirmation of these findings is
390: certainly required with integral field spectroscopy. %for instance through
391: %the Adaptive Optics assisted integral field spectrometer SINFONI which
392: %will soon be available at VLT. 
393: Another source of uncertainty is the possible star formation activity occurring in 
394: the host galaxy at radii larger than 250~pc; indeed the limited signal-to-noise 
395: ratio
396: of our spectra in the outer regions can only provide a relatively loose upper 
397: limit of 0.18 $\rm M_{\odot}~yr^{-1}~kpc^{-2}$ on the star formation rate per
398: unit surface area. Deeper observations are required to further constrain the
399: star formation in the host galaxy.
400: 
401: In the other two quasars the analysis of the line profile is more complex 
402: due to imperfect subtraction of some deep atmospheric absorption features 
403: and to a much lower Strehl ratio than in PG1126. 
404: There are some hints of a resolved narrow component of Pa$\alpha$, but need 
405: to be confirmed with higher quality spectra and higher Strehl Ratios.
406: 
407: \section{Conclusions}
408: 
409: We have obtained K-band spectroscopic observations assisted
410: by adaptive optics of three quasars at z$\sim$0.06
411: which are luminous in the far-IR. The adaptive optics correction
412: allows us the reach a diffraction limited angular resolution
413: ($\sim0.08''$, corresponding to $\sim100$ pc),
414: with Strehl ratios ranging from 5\% to 26\%.
415: In the quasar observed with the highest Strehl ratio (PG1126$-$041)
416: we spatially resolve a narrow component of Pa$\alpha$ on scales
417: of $\sim$50--100~pc. By comparing the spatial distribution of the
418: narrow Pa$\alpha$ with the distribution of the coronal line
419: [SiVI]1.96$\mu$m, we infer that most of the narrow Pa$\alpha$ is
420: due to star formation in the nuclear and circumnuclear region
421: of the quasar. We also derive that most of the far-IR emission
422: of this quasar is due to the nuclear star formation detected by us.
423: 
424: These observations clearly demonstrate that adaptive optics assisted
425: spectroscopy is a powerful tool to investigate the nuclear region
426: even in bright quasars.
427: 
428: 
429: \begin{acknowledgements}
430:       This work was partially supported by
431:       the Italian Ministry of Research (MIUR).
432:       We are grateful to the ESO staff on Paranal for having
433:       performed these observations in service mode.
434: \end{acknowledgements}
435: 
436: \begin{thebibliography}{}
437: 
438: \bibitem[2004]{alexander} Alexander, D.~M., 
439: Bauer, F.~E., Chapman, S.~C., et al.\ 
440: %Smail, I., Blain, A.~W., Brandt, W.~N., \& Ivison, R.~J., 
441: 2004, astro-ph/0401129
442: 
443: \bibitem[1999]{andreani01} 
444: Andreani, P., Franceschini, A., \& Granato, G.\ 1999, \mnras, 306, 161
445: 
446: \bibitem[2001]{canalizo01} Canalizo, G.~\& 
447: Stockton, A.\ 2001, \apj, 555, 719
448: 
449: \bibitem[2001]{cid01} Cid Fernandes, 
450: R., Heckman, T., Schmitt, H., Delgado, R.~M.~G., \& Storchi-Bergmann, T.\ 
451: 2001, \apj, 558, 81
452: 
453: \bibitem[2002]{courbin02} Courbin, F., et al.\ 
454: 2002, \aap, 394, 863
455: 
456: \bibitem[2003]{dunlop03} Dunlop, J.~S., et al.
457: McLure, R.~J., Kukula, M.~J., Baum, S.~A., O'Dea, C.~P., \& Hughes, D.~H.\ 
458: 2003, \mnras, 340, 1095
459: 
460: \bibitem[1995]{giannuzzo95} Giannuzzo, 
461: E., Rieke, G.~H., \& Rieke, M.~J.\ 1995, \apjl, 446, L5
462: 
463: \bibitem[1997]{heckman97} Heckman, T.~M.,  
464: Gonzalez-Delgado, R., Leitherer, C., et al.\ %Meurer, G.~R., Krolik, J., Wilson, 
465: %A.~S., Koratkar, A., \& Kinney, A.\, 
466: 1997, \apj, 482, 114
467: 
468: \bibitem[2003]{jahnke03} Jahnke, K.~\& 
469: Wisotzki, L.\ 2003, \mnras, 346, 304
470: 
471: \bibitem[1998]{kennicutt98} Kennicutt, R.~C.\ 1998, 
472: \araa, 36, 189
473: 
474: \bibitem[1995]{maiolino95} 
475: Maiolino, R., Ruiz, M., Rieke, G.~H., \& Keller, L.~D.\ 1995, \apj, 446, 
476: 561
477: 
478: \bibitem[1996]{maiolino96} Maiolino, 
479: R., Rieke, G.~H., \& Rieke, M.~J.\ 1996, \aj, 111, 537
480: 
481: \bibitem[1998]{maiolino98} 
482: Maiolino, R., Krabbe, A., Thatte, N., \& Genzel, R.\ 1998, \apj, 493, 650
483: 
484: \bibitem[2000]{maiolino00} Maiolino, R.   
485: Alonso-Herrero, A., Anders, S., et al.\ %Quillen, A., Rieke, M.~J., Rieke, G.~H., \& 
486: %Tacconi-Garman, L.~E.\ 
487: 2000, \apj, 531, 219
488: 
489: \bibitem[1996]{marconi96} Marconi, A., van der Werf, P.~P., 
490: Moorwood, A.~F.~M., \& Oliva, E.\ 1996, \aap, 315, 335
491: 
492: \bibitem[1994]{oliva94} 
493: Oliva, E., Salvati, M., Moorwood, A.~F.~M., \& Marconi, A.\ 1994, \aap, 
494: 288, 457
495: 
496: \bibitem[1999]{oliva99} 
497: Oliva, E., Origlia, L., Maiolino, R., \& Moorwood, A.~F.~M.\ 1999, \aap, 
498: 350, 9
499: 
500: \bibitem[2003]{reunanen03} 
501: Reunanen, J., Kotilainen, J.~K., \& Prieto, M.~A.\ 2003, \mnras, 343, 192
502: 
503: \bibitem[2000]{schade00} Schade, 
504: D.~J., Boyle, B.~J., \& Letawsky, M.\ 2000, \mnras, 315, 498
505: 
506: \bibitem[2001]{thompson01} Thompson, R.~I., Chary, R., Corbin, M.~R., \& 
507: Epps, H.\ 2001, \apjl, 558, L97 
508: 
509: \end{thebibliography}
510: 
511: \end{document}
512: