3 Observations and data reduction

Low dispersion spectra of 117 galaxies were obtained in several observing runs since 1995 using the imaging spectrographs BFOSC and LFOSC attached to the Cassini 1.5 m telescope (Loi) at Loiano (Italy) and the 2.1 m telescope (Can) of the Guillermo Haro Observatory at Cananea (Mexico), respectively.

  

Table 1: The spectrograph characteristics

Table 1 lists the characteristics of the two spectrographs in the adopted configurations:

The observations at Loiano were performed using a 2.0 or 2.5 arcsec slit, depending on the seeing conditions, oriented E-W. The exposure time ranged between 20 and 90 min according to the brightness of the target object. Observations at Cananea were carried out with a 3.1 arcsec fixed slit, oriented N-S. Every galaxy spectrum was preceded and followed by an exposure of a HeAr lamp (Loiano) or XeNe lamp (Cananea) to secure the wavelength calibration.

Data reduction was performed in the IRAF-PROS environment. After bias subtraction, when 3 or more frames of the same target were obtained, these were combined (after spatial alignment) using a median filter to help cosmic rays removal. Otherwise the cosmic rays were removed under visual inspection. The wavelength calibration was checked on known sky lines. These were found within $\sim\! 1$ Å from their nominal value, providing an estimate of the systematic uncertainty on the derived velocities of $\sim\! 50$ $\rm km\,s^{-1}\!$. After subtraction of the sky background, one-dimensional spectra were extracted from the frames. These spectra were analyzed with either of two methods:

1) individual line measurement: all spectra taken at Cananea and those obtained at Loiano prior to 1996 were inspected and emission/absorption lines were identified. Emission lines include H$\alpha$, N[II] and S[II]. Absorption lines include the Mg[I], CaFe and Na. The galaxy redshift was obtained from these individual measurements. If more than one line was identified, the galaxy redshift was derived as the weighted mean of the individual measurements, with weights proportional to the line intensities.

2) Cross correlation technique: spectra obtained at Loiano after 1996 were analyzed using the cross-correlation technique of Tonry & Davis (1979). This method is based on a "comparison" between the spectrum of a galaxy whose redshift is to be determined, and a fiducial spectral template of a galaxy (or star) of appropriate spectral type to contain the wanted absorption/emission lines. The basic assumption behind this method is that the spectrum of a galaxy is well approximated by the spectrum of its stars, modified by the effects of the stellar motions inside the galaxy and by the systemic redshift. For this purpose high signal-to-noise spectra were taken of two template galaxies: M105 (absorption lines) and NGC 1330 (emission lines), whose redshifts are 866 and 1039 $\rm km\,s^{-1}$ respectively.

The observed redshifts ($V_{\rm obs}$) were first transformed to Heliocentric ($V_{\rm hel}$), then corrected for the motion of our galaxy relative to the Cosmic Microwave Background ($V_{\rm CMB}$) according to Kogut et al. (1993).