1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass{emulateapj}
3:
4: %\usepackage{amsmath}
5: %\usepackage[]{natbib}
6:
7: \newcommand{\kms}{\>{\rm km}\,{\rm s}^{-1}}
8: \newcommand{\kmskpc}{\>{\rm km}\,{\rm s}^{-1}\,{\rm kpc}^{-1}}
9: \newcommand{\pc}{\>{\rm pc}}
10: \newcommand{\kpc}{\>{\rm kpc}}
11: \newcommand{\gyr}{\>{\rm Gyr}}
12: \newcommand{\msun}{\>{\rm M_{\odot}}}
13: \newcommand{\lsun}{\>{\rm L_{\odot}}}
14: \newcommand\degrees{^\circ}
15: \newcommand{\len}{a_{\rm B}}
16: \newcommand{\rbp}{R_{\rm B/P}}
17:
18: \def\etal{{et al.}}
19: \def\eg{{\it e.g.}}
20: \def\etc{{\it etc.}}
21: \def\ie{{\it i.e.}}
22: \def\cf{{\it cf.}}
23:
24: %%%%% AUTHORS - PLACE YOUR OWN MACROS HERE %%%%%
25:
26: %\input{figure_table_commands}
27:
28:
29: \slugcomment{Draft version on \today}
30: \shorttitle{Confirmation of a Kinematic Diagnostic for Face-On B/P Bulges}
31: \shortauthors{M\'endez-Abreu \etal}
32:
33: \begin{document}
34:
35: \title{Confirmation of a Kinematic Diagnostic for Face-On Box/Peanut-Shaped Bulges\altaffilmark{1}}
36: \author{J. M\'endez-Abreu \altaffilmark{2,3,4}, E. M. Corsini\altaffilmark{3},
37: Victor P. Debattista\altaffilmark{5,6}, S. De Rijcke\altaffilmark{7},
38: J. A. L. Aguerri\altaffilmark{8}, \& A. Pizzella\altaffilmark{3}}
39:
40: \altaffiltext{1}{Based on observations collected at the European
41: Southern Observatory, Chile (ESO N$^o$ 76.B-0324(A)).}
42: %
43: \altaffiltext{2}{INAF-Osservatorio Astronomico di Padova,
44: vicolo dell'Osservatorio 5, I-35122 Padova, Italy;
45: jairo.mendez@oapd.inaf.it}
46: %
47: \altaffiltext{3}{Dipartimento Astronomia, Universit\`a di Padova,
48: vicolo dell'Osservatorio 3, I-35122 Padova, Italy;
49: enricomaria.corsini@unipd.it;alessandro.pizzella@unipd.it}
50: %
51: \altaffiltext{4}{Universidad de La Laguna, Av. Astrof\'{\i}sico
52: Francisco S\'anchez s/n, E-38206 La Laguna, Spain}
53: %
54: \altaffiltext{5}{Centre for Astrophysics, University of Central Lancashire,
55: Preston, PR1 2HE, UK;
56: vpdebattista@uclan.ac.uk}
57: %
58: \altaffiltext{6}{RCUK Fellow}
59: %
60: \altaffiltext{7}{Sterrenkundig Observatorium, Universiteit Gent,
61: Krijgslaan 281, S9, B-9000 Gent, Belgium;
62: sven.derijcke@ugent.be}
63: %
64: \altaffiltext{8}{Instituto de Astrof\'\i sica de Canarias, Calle Via L\'actea s/n,
65: E-38200 La Laguna, Spain;
66: jalfonso@iac.es}
67:
68: %\date{{\it Draft version on \today}}
69:
70: %
71: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
72: %
73:
74: \begin{abstract}
75:
76: We present the results of high resolution absorption-line spectroscopy
77: of 3 face-on galaxies, NGC~98, NGC~600, and NGC~1703 with the aim of
78: searching for box/peanut (B/P)-shaped bulges. These observations test
79: and confirm, for the first, time the prediction that face-on
80: B/P-shaped bulges can be recognized by two minima in the profile along
81: the bar's major axis of the fourth Gauss-Hermite moment $h_4$ of the
82: line-of-sight velocity distribution (LOSVD). In NGC~98, a clear double
83: minimum in $h_4$ is present along the major axis of the bar and before
84: the end of the bar, as predicted. In contrast, in NGC~600, which is
85: also a barred galaxy but lacks a substantial bulge, we do not find any
86: significant kinematic signature for a B/P-shaped bulge. In NGC~1703,
87: which is an unbarred control galaxy, we found no evidence of a B/P
88: bulge.
89: %
90: We also show directly that the LOSVD is broader at the location of the
91: $h_4$ minimum in NGC~98 than elsewhere. This more direct method avoids
92: possible artifacts associated with the degeneracy between the
93: measurement of line-of-sight velocity dispersion and $h_4$.
94:
95: \end{abstract}
96:
97: \keywords{galaxies: bulges --- galaxies: evolution --- galaxies:
98: formation --- galaxies: kinematics and dynamics --- galaxies: spiral}
99:
100: %
101: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
102: %
103:
104: \section{Introduction}
105: \label{sec:intro}
106:
107: Roughly one-quarter of the visible light emitted by stars in the local
108: universe comes out of the bulges of disk galaxies \citep{per_sal_92,
109: fuk_etal_98}. Understanding how bulges form is therefore of great
110: importance to developing a complete picture of galaxy formation. The
111: processes by which bulges form are still debated. On the one hand,
112: the merger of dwarf-sized galactic subunits has been suggested as the
113: main path for bulge formation \citep{kau_etal_93}, which is supported
114: by the relatively homogeneous bulge stellar populations of the Milky
115: Way and M31 \citep{ferreras_etal_03, zoc_etal_03, ste_etal_03}.
116: Bulges formed in such mergers are termed `classical' bulges.
117: Alternatively, bulges may form via internal `secular' processes such
118: as bar-driven gas inflows, bending instabilities, and clump
119: instabilities \citep{com_san_81, pfenni_84, com_etal_90, pfe_fri_91,
120: rah_etal_91, noguch_99, imm_etal_04, athana_05,
121: deb_etal_06}. Moreover, the growth of the bulge out of disk material
122: may also be externally triggered by satellite accretion during minor
123: merging events \citep{agu_etal_01, eli_etal_06} or also via external
124: `secular' processes. Evidence for secular bulge formation includes
125: the near-exponential bulge light profiles \citep{and_san_94,
126: and_etal_95, cou_etal_96, dejong_96, car_etal_98, pri_etal_01,
127: car_etal_01, caroll_99, mac_etal_03}, a correlation between bulge and
128: disk scale lengths \citep{dejong_96, mac_etal_03, agu_etal_05,
129: men_etal_08}, the similar colors of bulges and inner disks
130: \citep{ter_etal_94, pel_bal_96, cou_etal_96, car_etal_07}, substantial
131: bulge rotation \citep{kormen_93, kor_etal_02}, and the presence of
132: B/P-shaped bulges in $\sim 45\%$ of edge-on disk galaxies
133: \citep{lut_etal_00}. A review of secular `pseudo-bulge' formation and
134: evidence for it can be found in \citet{kor_ken_04}. Standard cold
135: dark matter cosmology predicts that galaxies without classical bulges
136: should be rare \citep{don_bur_04}. Not only are they not rare
137: \citep[e.g.,][]{lau_etal_07}, they are more prevalent among normal
138: galaxies than classical bulges, exacerbating the disagreement between
139: theory and observations \citep{deb_etal_06}. It is therefore
140: important to determine which bulges are of the classical versus pseudo
141: variety, and which are a mix of both.
142:
143: $N$-body simulations show that barred galaxies have a tendency to
144: develop B/P bulges \citep{com_san_81, com_etal_90, pfe_fri_91,
145: rah_etal_91, deb_etal_06}. The fraction of edge-on galaxies with B/P
146: bulges is comparable to the fraction of disk galaxies containing bars
147: \citep[$\sim 60\%$][]{esk_etal_00, mar_jog_07} hinting that the two
148: are related. The presence of bars in edge-on galaxies with B/P bulges
149: has been established by the kinematics of gas and stars
150: \citep{kui_mer_95, mer_kui_99, bur_fre_99, bur_ath_99, bur_ath_05,
151: veg_etal_97, ath_bur_99, chu_bur_04}. However, the degeneracy
152: inherent in deprojecting edge-on galaxies makes it difficult to study
153: other properties of the host galaxy. For example, simulations show
154: that a bar can produce a B/P shape even if a massive classical bulge
155: formed before the disk \citep{ath_mis_02, deb_etal_05}. Understanding
156: the relative importance of classical and pseudo-bulges requires an
157: attempt at a cleaner separation of bulges, bars and peanuts, which is
158: easiest to accomplish in less inclined systems. At least in two cases,
159: NGC~4442 \citep{bet_gal_94} and NGC~7582 \citep{qui_etal_97}, a B/P
160: bulge can be recognized at inclination $i < 75\degrees$, which permits
161: also a direct identification of the bar. \citet{deb_etal_05} proposed
162: a kinematic diagnostic of B/P bulges which properly works up to a disk
163: inclination of about $30^\circ$ namely a double minimum in the fourth
164: moment, $h_4$, of the line-of-sight velocity distribution (LOSVD),
165: along the major-axis of the bar. These minima occur because at the
166: location of the B/P shape, the vertical density distribution of stars
167: becomes broader, which leads to a double minimum in $z_4$, the
168: fourth-order Gauss-Hermite moment of the vertical density
169: distribution. The kinematic moment $h_4$ is then found to be an
170: excellent proxy for the unobservable $z_4$. In contrast, the increase
171: in the vertical scale-height does not produce any distinct signature
172: of a B/P bulge and the vertical velocity dispersion, $\sigma_z$, is
173: too strongly dependent on the radial density variation to provide a
174: useful B/P bulge diagnostic. \citet{deb_etal_06} showed that the
175: diagnostic continues to hold even when gas is present since this sinks
176: to a radius smaller than that of the B/P bulge.
177:
178: In this Letter we present the first high resolution stellar kinematics
179: of face-on galaxies with the goal of searching for B/P-shaped bulges
180: and testing the diagnostic proposed by \citet{deb_etal_05}. In \S
181: \ref{sec:obsred} we describe the observations of the sample and data
182: reduction, while our results are presented in \S
183: \ref{sec:res}. Finally, the conclusions are given in \S
184: \ref{sec:conc}.
185:
186:
187: \section{Observations and data reduction}
188: \label{sec:obsred}
189:
190: \subsection{Sample selection}
191:
192: The barred galaxies NGC~98 and NGC~600 were selected in the
193: NASA/IPAC Extragalactic Database (NED) as bright ($B_T < 14$) and
194: undisturbed objects, with a disk inclination lower than $30^\circ$ and
195: a bar length larger than $10''$ to efficiently test the kinematic
196: diagnostic, a disk diameter smaller than $4'$ to allow an accurate sky
197: subtraction, no strong evidence of dust and no bright foreground stars
198: in the Digitized Sky Survey image. To select the sample galaxies,
199: their disk inclination, bar length, and bar position angle were first
200: determined from ellipse fits to the 2MASS $J-$band archival images.
201: The main properties of the sample galaxies are given in Table
202: \ref{tab:gals}.
203:
204: \placetable{table1}%Tabone
205:
206: \subsection{Long-slit spectroscopy}
207: \label{sec:spectroscopy}
208:
209: The spectroscopic observations were carried out in service mode at the
210: Very Large Telescope (VLT) at the European Southern Observatory (ESO)
211: on 2005 October 2-9 and November 2.
212: %
213: The Focal Reducer Low Dispersion Spectrograph 2 (FORS2) mounted the
214: volume-phased holographic grism GRIS\_1028z$+$29 with 1028 $\rm
215: grooves\;mm^{-1}$ and the $0\farcs7\,\times\,6\farcm8$ slit.
216: %
217: The detector was a mosaic of 2 MIT/LL CCDs, each with $2048\times2068$
218: pixels of $15\,\times\,15$ $\mu$m$^2$. The wavelength range from 7681
219: to 9423 \AA\ was covered in the reduced spectra with a reciprocal
220: dispersion of 0.858 \AA\ pixel$^{-1}$ and a spatial scale of
221: $0\farcs250$ pixel$^{-1}$ after a $2\times2$ pixel binning.
222: %
223: The spectra were taken along the bar major axis of NGC~98 ($\rm
224: P.A.=32\fdg2$) and NGC~600 ($\rm P.A. = 22\fdg1$) and along the disk
225: major axis of NGC~1703 ($\rm P.A. = 148\fdg8$). The total integration
226: time for each galaxy was 3 hours, in four exposures of 45 minutes
227: each.
228: %
229: Comparison lamp exposures obtained for each observing night ensured
230: accurate wavelength calibrations. Spectra of G and K giant stars
231: served as kinematic templates. The average seeing FWHM was $1\farcs3$
232: for NGC~98, $0\farcs9$ for NGC~600, and $0\farcs9$ for NGC~1703 as
233: measured from the ESO Differential Imaging Meteo Monitor.
234:
235: Using standard IRAF\footnote{IRAF is distributed by NOAO, which is
236: operated by AURA Inc., under contract with the National Science
237: Foundation.} routines, all the spectra were bias subtracted,
238: flat-field corrected, cleaned of cosmic rays, corrected for bad
239: pixels, and wavelength calibrated as in Debattista et al. (2002). The
240: accuracy of the wavelength rebinning ($\approx1 \kms$) was checked by
241: measuring wavelengths of the brightest night-sky emission lines. The
242: instrumental resolution was $1.84\pm0.01$ \AA\ (FWHM) corresponding to
243: $\sigma_{\rm inst} = 27$ $\kms$ at 8552 \AA .
244: %
245: The spectra obtained for the same galaxy were co-added using the
246: center of the stellar continuum as reference. In the resulting spectra
247: the sky contribution was determined by interpolating along the
248: outermost $\approx30''$ at the edges of the slit and then subtracted.
249:
250:
251: \subsection{Photometry}
252:
253: We analyzed the uncalibrated acquisition images from the VLT to derive
254: the photometric properties of the sample galaxies.
255: %
256: Isophote-fitting with ellipses, after masking foreground stars and bad
257: pixels, was carried out using the IRAF task ELLIPSE.
258: %
259: Under the assumption that the outer disks are circular, their
260: inclination was determined by averaging the outer isophotes. All 3
261: galaxies have $i < 30\degrees$ (Table~\ref{tab:gals}).
262: %
263: The semi-major axis length, $\len$, of the bars of NGC~98 and NGC~600
264: (Table~\ref{tab:gals}) was measured from a Fourier decomposition as in
265: \citet{agu_etal_00}. It was calculated using the bar/interbar
266: intensity ratio $I_{\rm b}/I_{\rm ib}$. The bar intensity is defined
267: as the sum of the even Fourier components, $I_{0}+I_{2}+I_{4}+I_{6}$,
268: while the inter-bar intensity is given by $I_{0}-I_{2}+I_{4}-I_{6}$.
269: The bar length is the FWHM of radial profile of $I_{\rm b}/I_{\rm
270: ib}$. This method was applied by \citet{ath_mis_02} to analytic models
271: demonstrating its accuracy in measuring $\len$.
272:
273:
274: %\placeFigone
275: \placefigure{f1}
276:
277: \subsection{Kinematics}
278:
279: The stellar kinematics of the three galaxies were measured from the
280: galaxy absorption features present in the wavelength range centered on
281: the Ca~{\small II} triplet ($\lambda\lambda\,8498,\,8542,\,8662\,$\AA
282: ) using the Penalized Pixel Fitting method
283: \citep[pPXF,][]{cap_ems_04}. The spectra were rebinned along the
284: dispersion direction to a logarithmic scale, and along the spatial
285: direction to obtain a signal-to-noise ratio $S/N \ga 40$
286: \AA$^{-1}$. It decreases to $S/N=20$ \AA$^{-1}$ at the outermost
287: radii.
288:
289: A linear combination of the template stellar spectra, convolved with
290: the line-of-sight velocity distribution (LOSVD) described by the
291: Gauss-Hermite expansion by \citet{vdm_fra_93} was fitted to each
292: galaxy spectrum by $\chi^2$ minimization in pixel space. This allowed
293: us to derive profiles of the mean velocity ($v_{\rm los}$), velocity
294: dispersion ($\sigma_{\rm los}$), third ($h_3$), and fourth-order
295: ($h_4$) Gauss-Hermite moments. The uncertainties on the kinematic
296: parameters were estimated by Monte Carlo simulations with photon, read
297: out and sky noise.
298: %
299: Extensive testing on simulated galaxy spectra was performed to provide
300: an estimate of the biases of the pPXF method with the adopted
301: instrumental setup and spectral sampling. The simulated spectra were
302: obtained by convolving the template spectra with a LOSVD parametrized
303: as a Gauss-Hermite series and measured as if they were real. No bias
304: was found in the ranges of $S/N$ and $\sigma_{\rm los}$ which
305: characterize the spectra of the sample galaxies. The values of $h_3$
306: and $h_4$ measured for the simulated spectra differ from the intrinsic
307: ones only within the measured errors \citep[see also][]{ems_etal_04}.
308:
309:
310: \section{Results}
311: \label{sec:res}
312:
313: The photometric and kinematic profiles of the sample of galaxies are
314: shown in Figure~\ref{fig:kine}. We focus here on the $h_4$ profiles; a
315: full analysis of this data set will be presented in a forthcoming
316: paper. For the unbarred galaxy, NGC~1703, the $h_4$ profile shows no
317: sign of a minimum. The same is true for the almost bulgeless, barred
318: NGC~600. In both these galaxies $\sigma > 30 \kms$, including within
319: the bar in NGC~600. Thus, our failure to find an $h_4$ minimum cannot
320: be ascribed to low spectral resolution. In NGC~98, instead, we find a
321: double minimum in the $h_4$ profile. The two minima are symmetric with
322: respect to the galaxy center ($|r|\approx5''$). In the simulations of
323: \citet{deb_etal_05} the minima in $h_4$ are $\sim 0.05 - 0.1$ deep,
324: comparable to the minimum in the profile of NGC~98.
325:
326: We confirmed that our kinematics, and minima in $h_4$, are not a
327: result of details of the pPXF method by repeating the analysis using
328: the Fourier Correlation Quotient (FCQ) method \citep{bender_90} as
329: done in \citet{piz_etal_08}. Nonetheless expansions in Gauss-Hermite
330: moments are degenerate \citep{gerhar_93, vdm_fra_93} since, for the
331: even moments, it is possible to obtain comparably good fits by trading
332: $\sigma_{\rm los}$ for $h_4$. In order to confirm that the signature
333: of a B/P bulge obtained in NGC~98 is not due to a conspiracy of such
334: degeneracies we introduced an LOSVD broadening measure defined for
335: each radius as
336: %
337: \begin{equation}
338: {\cal B}(x_1,x_2) \equiv W(x_2)/W(x_1),
339: \end{equation}
340: %
341: where
342: %
343: \begin{equation}
344: W(x) = \int_{v^{\rm min}_{\rm los}}^{v^{\rm max}_{\rm los}} F(v)dv,
345: \end{equation}
346: %
347: with $F(v)$ the cumulative distribution of the LOSVD. The values of
348: $v^{\rm min}_{\rm los}$ and $v^{\rm max}_{\rm los}$ correspond to
349: $F(v^{\rm min}_{\rm los}) = 0.5\,(1-x)$ and $F(v^{\rm max}_{\rm los})
350: = 0.5\,(1+x)$, respectively.
351: %
352: ${\cal B}(x_1,x_2)$ has the advantage of not being sensitive to the
353: wings of the distribution, unlike the kurtosis and is a non-parametric
354: measure of the LOSVD shape. Experiments with $N$-body simulations,
355: including those presented in \citet{deb_etal_05}, showed that ${\cal
356: B}(0.7,0.9)$ is a useful peanut diagnostic in the same spirit as
357: $h_4$. Figure \ref{fig:kine} plots ${\cal B}(0.7,0.9)$. A clear peak
358: in the broadening is apparent at $|r| \simeq 5\arcsec$ in NGC~98. We
359: conclude that the evidence for a B/P-shaped bulge in NGC~98 is robust.
360:
361:
362: \section{Conclusions}
363: \label{sec:conc}
364:
365: We have identified a B/P-shaped bulge in one galaxy from a sample of
366: three face-on galaxies. In the unbarred control galaxy NGC~1703 we
367: had not expected to find a B/P bulge. The failure to find a
368: double-minimum in $h_4$ is therefore fully consistent with previous
369: results \citep{chu_bur_04}. Of the two barred galaxies, NGC~98 has
370: clear evidence of a B/P bulge while NGC~600 does not. The absence of
371: a B/P shape in the latter galaxy is not surprising since it appears to
372: not have a bulge and its bar has a lumpy structure.
373:
374: If we identify the radius of the B/P bulge, $\rbp$, with the location
375: of the minimum in $h_4$, as found in simulations \citep{deb_etal_05},
376: then we find $\rbp \simeq 0.35 \len$, where $\len$ is the bar
377: semi-major axis. Similarly, \citet{kor_ken_04} noted that the maximum
378: radius of the boxy bulge is about one-third of the bar radius.
379: Simulations also produce B/P-bulges which are generally smaller than
380: the bar \citep{she_sel_04, deb_etal_05}.
381:
382: While observations have established the presence of bars in the
383: majority of systems with B/P-bulges, the converse has not been
384: established. \citet{deb_etal_05} demonstrate how the presence of a
385: classical bulge can in some cases mask the presence of B/P bulge.
386: Moreover, the ratio of radii of bulges and B/P bulges is determined by
387: resonances. For these various reasons it would be very instructive to
388: repeat measurements such as here for a sample of barred galaxies.
389:
390:
391: %\bigskip
392: %\noindent
393: %{\bf Acknowledgments.}
394:
395: %\noindent
396: %
397: \acknowledgements We thank Eva Grebel for her support and Lorenzo Morelli for
398: fruitful discussion.
399: %
400: E.M.C. and V.P.D. thank the Instituto de Astrof\'\i sica de Canarias for
401: hospitality during part of this project.
402: %
403: J.M.A. acknowledges support from the Istituto nazionale di Astrofisica (INAF).
404: %
405: E.M.C. and A.P. receive support from the grant CPDA068415/06 by Padua University.
406: %
407: S.D.R. acknowledges Postdoctoral Fellowship support from the Fund for
408: Scientific Research--Flanders, Belgium (FWO).
409: %
410: J.A.L.A. is funded by the Spanish DGES, grant AYA-2007-67965-C03-01.
411: %
412: This research has made use of the NASA/IPAC Extragalactic Database,
413: Digitized Sky Survey, and 2 Micron All Sky Survey.
414:
415: %%%%%%%%%%%%%%%
416: % Reference List
417: %%%%%%%%%%%%%%%
418:
419: \bibliographystyle{aj.bst}
420: \bibliography{allrefs}
421:
422: \clearpage
423:
424:
425: %\newcommand{\placeTabone}{
426: \begin{deluxetable}{llcccrc}
427: \tablecolumns{7}
428: \tablewidth{0pc}
429: \tablecaption{Parameters of the sample galaxies.}
430: \tablehead{
431: \multicolumn{1}{c}{Object} &
432: \multicolumn{1}{c}{Type} &
433: \multicolumn{1}{c}{$B_T$} &
434: \multicolumn{1}{c}{$D_{25}$} &
435: \multicolumn{1}{c}{$M^0_{B_{T}}$} &
436: \multicolumn{1}{c}{$i$} &
437: \multicolumn{1}{c}{$a_B$}}
438: \startdata
439: NGC~98 & SBbc & 13.4 & $1\farcm7$ & $-22.4$ & $26^\circ$ & $14\arcsec \pm 1\arcsec$ \\
440: NGC~600 & SBd & 12.9 & $3\farcm3$ & $-20.0$ & $21^\circ$ & $12\arcsec \pm 1\arcsec$ \\
441: NGC~1703 & Sc & 11.9 & $3\farcm0$ & $-20.6$ & $15^\circ$ & \nodata
442: \enddata
443: \tablecomments{Hubble types, apparent magnitudes, and diameters
444: are from the NASA/IPAC Extragalactic Database (NED). Absolute
445: magnitudes are calculated from $B^0_T$ (NED) with distance from
446: $V_{\rm CMB}$ (NED) and $H_0=75$ $\kmskpc$. Disk inclinations and bar
447: semi-major axis lengths are from this paper.}
448: \label{tab:gals}
449: \end{deluxetable}
450:
451: %\placeFigone
452: \clearpage
453:
454: \begin{figure*}[ht!]
455: %\vspace*{3truecm}
456: \begin{center}%$
457: %\begin{array}{ccc}
458: \includegraphics[width=2.2in, bb=110 350 460 1100]{f1a.ps}% &
459: \includegraphics[width=2.2in, bb=110 350 460 1100]{f1b.ps}% &
460: \includegraphics[width=2.2in, bb=110 350 460 1100]{f1c.ps}
461: %\end{array}$
462: \end{center}
463: %\vspace*{-1truecm}
464: \caption{Morphology and stellar kinematics of NGC~98 (left panels),
465: NGC~600 (central panels), and NGC~1703 (right panels).
466: %
467: For each galaxy the top panel shows the VLT/FORS2 $I-$band image.
468: The slit position and image orientation are indicated. The inset
469: shows the portion of the galaxy image marked with a white box. The
470: gray scale and isophotes were chosen to enhance the features observed
471: in the central regions.
472: %
473: The remaining panels show from top to bottom the radial profiles of
474: surface brightness (extracted along the slit with an arbitrary zero
475: point), ${\cal B}(0.7,0.9)$, $h_4$, $h_3$, $\sigma_{\rm los}$ (in
476: $\kms$ ), and velocity $v$ (in $\kms$ , obtained by subtracting the
477: systemic velocity from $v_{\rm los}$). The two vertical lines
478: indicate the location of the $h_4$ minima in NGC~98.}
479: \label{fig:kine}
480: \end{figure*}
481:
482: %%%%%%%%%%%%%%%
483: % End of Document
484: %%%%%%%%%%%%%%%
485:
486: \end{document}
487: