2 Observations and data reduction

2.1 Galaxy sample

All the observed galaxies (Table 1) belong to sample of 35 E and S0 galaxies of the Coma cluster studied in Paper I). For details about their morphological classification and relevant photometric properties (i.e. total magnitude, effective radius, mean surface brightness within effective radius, ellipticity at effective radius and luminosity weighted a₄ parameter) the reader is referred to that paper. They were selected from the sample of Paper I as the objects with the most extended and precise major axis kinematics and therefore best suited for dynamical modelling, balancing the number of E and S0 types.

2.2 Long-slit spectroscopy

Long-slit spectroscopic data of the sample galaxies were obtained with the 2.4-m Hiltner telescope of the MDM Observatory at Kitt Peak, Arizona, USA, during four different runs between 1999 and 2001. Details of the instrumental set-up of the observations carried out on 16-20 April 1999 (run 1), 23-26 April 1999 (run 2), 26-30 May 2000 (run 3) and 9-12 February 2001 (run 4) are given in Table 2.

   

Table 2: Instrumental set-up of spectroscopic observations.

Parameter	Run 1-2	Run 3-4
Spectrograph	Modular	Modular
Grating	1200 $\rm gr~mm^{-1}$	1200 $\rm gr~mm^{-1}$
CCD	"Charlotte'' SITe	"Echelle'' SITe
Pixel number	$1024\times1024$	$2048\times2048$
Pixel size	$24\times24$ $\rm\mu m^2$	$24\times24$ $\rm\mu m^2$
Gain	3.2 $e^{-}{\rm ADU}^{-1}$	2.7 $e^{-}{\rm ADU}^{-1}$
RON	5.5 e-	7.9 e-
Scale	$\rm0\hbox{$.\!\!^{\prime\prime}$ }606\;pixel^{-1}$	$\rm0\hbox{$.\!\!^{\prime\prime}$ }606\;pixel^{-1}$
Dispersion	$\rm 1.20\;\AA\;pixel^{-1}$	$\rm 1.01\;\AA\;pixel^{-1}$
Slit width	$2\hbox{$.\!\!^{\prime\prime}$ }0$	$1\hbox{$.\!\!^{\prime\prime}$ }9$
Wavelength range	4916-5881 Å	4585-6470 Å
Instrumental FWHM	3.20 Å	3.11 Å
Instrumental $\sigma^{\rm a}$	79 $\rm km~s^{-1}$	77 $\rm km~s^{-1}$
Seeing FWHM	$1\hbox{$.\!\!^{\prime\prime}$ }9$- $3\hbox{$.\!\!^{\prime\prime}$ }0$	$2\hbox{$.\!\!^{\prime\prime}$ }0$- $3\hbox{$.\!\!^{\prime\prime}$ }0$

$^{\rm a}$ Derived at the 5170 Å Mg triplet.

Minor-axis spectra were obtained for all the sample galaxies, while offset spectra with the slit parallel to the major axis were obtained only for GMP 0756, GMP 1176, GMP 1990. GMP 3818 was observed along a diagonal axis. In addition, we took spectra of GMP 1176 and GMP 3792 along their major axes to perform a consistency check with measurements of kinematics and line strength indices of Paper I. The typical integration time of the galaxy spectra was 3600 s. Total integration times and slit position angle of the galaxy spectra as well as the log of the spectroscopic observations are given in Table 1. At the beginning of each exposure the galaxy was centered on the slit using the guiding camera. In each run several spectra of giant stars with spectral type ranging from late-G to early-K were obtained to be used as templates in measuring stellar kinematics and line strength indices. The template stars were selected from Faber et al. (1985) and González (1993). Additionally we observed at least one flux standard star per night to calibrate the flux of the spectra before the line indices were measured. Spectra of the comparison Hg-Ne-Ar-Xe arc lamp were taken before and/or after object exposures to allow an accurate wavelength calibration. The value of the seeing FWHM during the observing run as measured by fitting a two-dimensional Gaussian to the guide star is given in Table 2.

2.3 Basic data reduction

All the spectra were bias subtracted, flatfield corrected, cleaned of cosmic rays, corrected for bad columns and wavelength calibrated using standard MIDAS routines. The flatfield correction was performed by means of both quartz lamp and twilight sky spectra, which were normalized and divided into all the spectra, to correct for pixel-to-pixel sensitivity variations and large-scale illumination patterns due to slit vignetting. Cosmic rays were identified and eliminated by interpolating over as in Bender et al. (1994, BSG94 hereafter). The residual cosmic rays were eliminated by manually editing the spectra. The wavelength calibration was performed by means of the MIDAS package XLONG. Each spectrum was rebinned using the wavelength solution obtained from the corresponding arc-lamp spectrum. We checked that the wavelength rebinning had been done properly by measuring the difference between the measured and predicted wavelengths (Osterbrock et al. 1996) for the brightest night-sky emission lines in the observed spectral range. The resulting accuracy in the wavelength calibration is $\sim$0.08 Å corresponding to $\sim$5  $\rm km~s^{-1}$ at 5170 Å. The instrumental resolution was derived as the mean of the Gaussian FWHMs measured for a number of unblended arc-lamp lines which were distributed over the whole spectral range of a wavelength-calibrated spectrum. The mean FWHM of the arc-lamp lines and the corresponding instrumental resolution derived at 5170 Å is given in Table 2. In the galaxy and the stellar spectra the contribution of the sky was determined by interpolating along the outermost 10''-20'' at the two edges of the slit, where the galaxy or stellar light was negligible, and then subtracted. A sky subtraction better than $1\%$ was achieved. All the spectra were corrected for CCD misalignment following BSG94. The spectra obtained for the same galaxy along the same axis were coadded using the center of the stellar continuum as a reference. This allowed to improve the signal-to-noise ratio (S/N) of the resulting two-dimensional spectrum. A one-dimensional spectrum was obtained for each kinematical template star as well as for each flux standard star. The spectra of the kinematical Lick-system templates were deredshifted to laboratory wavelengths.