We defined three criteria for selecting a sample of galaxies likely to contain the highest number of starburst galaxies. The selected galaxies must possess:
The Markarian survey contains 1500 objects selected for their excess of blue
and UV emission among which about 1100 are galaxies. The catalog of Markarian
galaxies is the largest sample of active galaxies known so far. The excess of
blue luminosity and UV emission is the signature of either a high star
formation rate or the presence of an Active Galactic Nucleus (AGN) in the
central region of the galaxies. We extracted our sample from the extensive
compilation of Markarian galaxies by
Mazzarella & Balzano (1986)
which lists valuable
data, such as morphology, magnitude, radial velocity, radio (6, 11 and 21 cm
continuum) and infrared (60 
m) flux of the galaxies.
The very successful IRAS mission, which covered about 98% of the sky in
four bands centered at 12, 25, 60 and 100 m has led to valuable
information on the origin of the FIR radiation and its relation to the dust
content of star forming regions. Interstellar dust is heated by absorption of
a large fraction of UV radiation and this energy is reradiated
in the FIR range.
In most IRAS galaxies, the heating of interstellar dust is due to a high star
formation rate and/or an AGN which provides a sufficiently strong ionizing
flux to produce the observed FIR radiation. We thus selected those galaxies of
the Markarian catalog which were detected by the IRAS satellite, and listed in
two catalogs of IRAS sources, namely the Point Source Catalog (PSC) which
contains all the bright point sources and the Faint Source Catalog (FSC) which
contains fainter extended sources.
With the first two criteria (UV- and FIR-bright), we expected to select a sample of galaxies with a high proportion of starburst galaxies. Our last criterion was to select galaxies with a barred morphology, because of their peculiar dynamics. The bar is thought to enhance the flow of molecular gas toward the center of spiral galaxies where a high concentration of this gas will trigger nuclear starbursts. To select galaxies with bars, we used the morphological classifications provided by Mazzarella & Balzano (1986) and LEDA (Lyon-Meudon Extragalactic Database).
Finally, the combination of two activity criteria (Markarian + IRAS) and the presence of a bar defined a sample of 144 barred Markarian IRAS galaxies.
This sample differs from others used for studying the starburst phenomenon in galaxies. Indeed, many works in this field are based on samples of small irregular or blue compact dwarf galaxies like "HII galaxies'' (e.g. Terlevich et al.1991), or samples of highly luminous infrared galaxies (e.g. Veilleux et al.1995), or even very heterogeneous samples of galaxies of mixed types. Our sample is of course not free from selection effects; the first selection criterion excludes galaxies with little or no UV excess, the second one discriminates against weaker starbursts, with less dust and molecular gas, the third one favors late-type galaxies, and the Markarian catalog does not contain nearby star-forming galaxies (such as M 82 or NGC 253).
The global properties of the original sample are listed in Table 1. The first column gives the number of the galaxy in the Markarian catalog. Columns 2 and 3 indicate the equatorial coordinates (equinox 1950). Cross identifications with other catalogs are given in Col. 4. The morphological properties of the galaxies, based on the classification system of de Vaucouleurs et al. (1991, RC3), are given in Cols. 5 to 8. Column 6 indicates if the galaxy has ring structures and Col. 7 if it belongs to a multiple system. The inclination and heliocentric radial velocity are given in Cols. 9 and 10. The apparent blue magnitude, corrected for both Galactic and internal extinction with the method of de Vaucouleurs et al. (1991, RC3), and the absolute blue magnitude are reported in Cols. 11 and 12. All the above general properties come from LEDA.
The IRAS data, reported in Cols. 13 to 20 of Table 1, come
from the Faint Source Survey
(Bicay et al.1995). These data are
more accurate than those from the PSC or the FSC, and represent in fact the
latest version of the IRAS catalog. The flux densities (in Jy) at 12, 25, 60
and 100 m are reported in Cols. 14, 16, 18 and 20
respectively. A quality code is assigned to each flux density in order to
evaluate the reliability of the measurements. The code value is equal to 3 if
the measurement is of high quality, to 2 if the quality is moderate and to 1
if it corresponds to an upper limit (90% confidence level). No measurements
of the flux density at 100 m are reported with a quality code greater
than or equal to 3, because of the strong contamination by interstellar cirrus
at this wavelength. Table 1 is given in electronic form only.
These data will be used in Sect. 6 to derive the general properties of the SBNGs.
Table 1:
Global properties of the sample
of barred Markarian galaxies (available in electronic form only)
Copyright The European Southern Observatory (ESO)