Astron. Astrophys. 324, 877-887 (1997)

4. Conclusions

In this paper a study of the neutral hydrogen properties of a sample of 108 galaxies, based on short 21-cm observations with the WSRT, has been presented. The observations, described in Papers I and II, provide one-dimensional information on the kinematics and spatial distribution of the neutral hydrogen along the major axis of each galaxy in the form of a position-velocity (XV) map. The kinematics will be discussed in an accompanying paper (Rhee & Broeils 1996). From these maps a number of H I properties were derived, like H I fluxes and masses, radial H I surface density profiles, and isophotal diameters of the H I discs.

In the first part of this paper these properties were correlated with the optical properties of the underlying galaxies, like luminosities, morphological type and an optical diameter. The main results are:

• The H I diameter (), measured at a surface density level of , correlates strongly with the (inclination- and absorption-corrected) optical diameter (), defined at the isophotal level. The diameter ratio, , shows no dependence on type or luminosity; the average value is . None of the galaxies in our sample has an H I disc smaller than the stellar disc.
• The average H I surface density , defined by , hardly varies from galaxy to galaxy, although the H I masses vary by more than a factor 400. The average value is . The 's of early-type spirals in our sample are only slightly smaller than the average value for the whole sample, probably due to the fact that these galaxies were selected for their H I -richness.
• The ratio shows no correlation with luminosity, and only a weak correlation with type, maybe partly due to the above-mentioned selection effect. The mass ratio is more strongly correlated with morphological type, in the sense that this ratio increases for later types. This indicates that the neutral gas component becomes dynamically more important in late-type, gas-rich galaxies, which is confirmed by detailed mass models based on two-dimensional 21-cm synthesis observations (e.g. Carignan & Freeman 1985, Lake et al. 1990, Côté et al. 1991, Broeils 1992).
• The total mass inside correlates strongly with the morphological type of a galaxy, more strongly than H I mass or blue luminosity versus type. The strong correlation between total mass and luminosity is expected to be the basis of the small scatter in the Tully-Fisher relationship between luminosity and rotation velocity. The global mass-to-light ratios in the blue band () for the dwarf galaxies in our sample are slightly smaller than those for the more luminous galaxies. This indicates that one has to be careful with the use of dwarf galaxies in the Tully-Fisher relation.

The second part of this paper described a Principal Component Analysis of the radial H I density distributions of the sample galaxies. We showed that of the variations among surface density profiles can be accounted for by just two principal components. The first, dominant component is clearly linked to the "scale"; it has the same sign at all radii. In other words, the process that is linked to this component works in such a way that if it makes the surface density higher at one point it makes it higher at all other points. The second component is linked to the behaviour of the density profile in the central regions: peak or depression. The third component, which still accounts for of the variation, is most likely responsible for bumps and wiggles in the observed density profiles and could be linked to the presence of H I arms and rings. This could be confirmed by performing PCA to a large sample of galaxies with full two-dimensional synthesis observations of the neutral hydrogen distribution.

© European Southern Observatory (ESO) 1997

Online publication: May 5, 1998

helpdesk.link@springer.de