Annu. Rev. Astron. Astrophys. 1978. 16: 103-39 Copyright © 1978 by Annual Reviews. All rights reserved

2.2. Distribution of the Gas

Although it is not the main subject of this chapter, we review briefly the results obtained on the distribution of the gas, since they bear on the radial velocity measurements to be discussed later. Observations of the optical emission lines obviously require the presence of ionized gas, either in discrete HII regions, or, in favorable cases, in the more tenuous areas between them. Hodge (1974a) has reviewed the available information on the radial distribution of H II regions: the surface number-densities are small in the central areas of intermediate- and later-type galaxies, increase to a maximum, then decrease again to small values near the optical boundaries as seen on, for example, the Palomar Sky Survey. The deficiency of HII regions in the central areas is in general correlated with the relative decrease in the HI surface density there (Roberts 1972, Davies 1972), which supports the widely held view that star formation is less vigorous in regions of lower gas density. Oort (1974) and Shu (1974) have suggested that the central depression is a result of accelerated depletion of the gas by star formation due to more frequent passages of the gas through the density wave in the inner regions. A detailed application of this idea to the observations shows a general agreement for M51 (Shane 1975) and to a lesser extent for M81 (Segalovitz 1975).

For several spiral galaxies of large angular size the angular resolution of the radio HI synthesis maps is sufficient to separate spiral arms from each other. The situation for M31, M33, M51, M81, and M101 has been reviewed by Allen (1975a); in all these systems the HI distribution shows spiral structure associated with the optical morphology in a general way. In M51 the HI arms are slightly but systematically displaced from the H arms, coinciding instead with the dust lanes (see also Shane 1975), as is also the case for M81 (Rots 1975). The displacement has been interpreted in terms of migration of newly formed stars away from the region of the shock in the density-wave model for spiral-arm structure described by Roberts (1969).

The correlation of HI with HII becomes more complicated when one looks in greater detail: Boulesteix et al. (1974) have pointed out that although on the kiloparsec scale the HII region number densities correlate with regions of higher-than-average HI gas density in M33 (see also Israel & van der Kruit 1974), on a smaller scale one finds that most of the HII regions are in fact situated near the edges of dense HI concentrations. Emerson (1974) has found a similar situation in M31. He offers further the suggestion that the effect is only an apparent one, caused by obscuration of HII regions embedded within the HI clouds by the greater concentrations of dust probably present there. On the other hand, Allen (1975b) has called attention to the existence of a contrasting situation for the giant HII complexes in M101; although dust is known to be present within them (Israel et al. 1975), the optically bright complexes coincide with the brightest areas on the radio HI maps. The explanation probably involves a high degree of inhomogeneity of the dust, HI, and ionized gas in these giant complexes, and much detailed structure will most likely appear when the HI can be measured with even higher angular resolution. Baldwin (1976) noted that the radio maps of M33 and the LMC show no thermal sources that are not identified with an HII region and are at the same time coincident with a peak in the HI distribution; he concludes from this that no luminous HII regions can be hidden by dust.

Accurately calibrated maps of the distribution and motions of the ionized gas in spiral galaxies would provide the basis for quantitative comparison with radio-synthesis observations. As far, as we know the only example presently available is the H map of M51 presented by Tully (1974a) and used by Segalovitz (1976, 1977) and van der Kruit (1977a) in a new discussion of the radio continuum emission (see also van der Kruit 1978).