3.2. Observation of the Diffuse and Molecular Gas Content of Ring Galaxies

Silverglade and Krumm (1978) made Arecibo observations of eleven ring galaxies and detected six of them. The authors concluded that the HI masses of ring systems were not significantly different from those of "normal" spiral galaxies. Observations at Parkes of the Cartwheel by Mebold, Goss and Fosbury (1977) showed that it contained significant amounts of neutral hydrogen. In the Ph.D Thesis of Jeske (1986), 16 out of 28 ring galaxies were detected using the Arecibo radio telescope.

Very few high-resolution aperture synthesis HI observations have been made of classical ring galaxies. In a VLA study of NGC 2793 by Ghigo, Hine and van der Hulst (in preparation), the HI was found to be clumpy and concentrated mainly in the bright knotty parts of the ring. Kinematically, this galaxy, identified as ring galaxy number 6 of Thompson's (1977) list, shows a rising rotation curve and the kinematics of the ring are consistent with a large expansion velocity (see section 3.3). (We note that new observations at the VLA have recently been made (J. Higdon, personal communication) and these higher sensitivity observations will soon be available.)

The Cartwheel ring galaxy has been the subject of a very detailed and elegant HI study by Higdon (1993) as part of his Ph.D thesis. Figure 9 shows that neutral hydrogen emission is seen associated with the ring, with some additional material found both outside and inside the ring. The HI surface density is lower in the region of the bright HII regions in the southern quadrant of the ring, but has a clumpy distribution that roughly follows the outer ring. Gas is concentrated in regions of the ring with less active star formation and some gas is found outside the ring. Although there is no obvious major concentration of gas associated with the spokes, a faint component of HI emission is found inside the ring. In general though, very little HI is found inside the ring. The kinematic results from these observations have been used by Struck-Marcell and Higdon (1993) to constrain models of the mass distribution in the Cartwheel and as the basis on numerical models of the collision (see Section 6.3). As in the case of NGC 2793, a rising rotation curve was found.

Figure 9 also shows another important fact. The extended HI emission found outside the Cartwheel is very asymmetrically distributed in the Cartwheel group: all of it being found in the general direction of the companions to the NE. A faint filament was discovered by Higdon (1993) extending from the Cartwheel all the way to the most distant companion G3. The asymmetric nature of the HI relative to the Cartwheel is strongly suggestive of an impact from the south-west which has scattered debris into the northeast quadrant. This is further evidence that the Cartwheel was formed by a collision.

Figure 9. James Higdon's medium resolution (10.7 × 9.1 arcsecs2 FHWM) HI observations of the Integrated HI emission from the Cartwheel obtained using the combined B, C and D arrays of the VLA superimposed on a CTIO 4-meter prime-focus plate of the galaxy (Higdon 1993; Figure 3a of Ph.D thesis). Strong emission is seen associated with the ring and is stronger away from the regions of intense star formation. Fingers of HI also point inwards towards the center of the Cartwheel which is generally deficient in HI. Gas is also seen to the north-east associated with the late-type companion and the far northern companion. Much scattered low-surface-brightness emission is also seen in this region.

At the time of writing, a number of new HI studies are underway in both the northern and southern hemispheres. Work is being performed by J. Wallin (G.M.U.) and J. Higdon (N.R.A.O.) to study a number of southern ring galaxies including Vela and AM064-741. In the north VLA observations of VIIZw466 and Arp 10 have been made by V. Charmandaris and one of us (PNA). HI emission was detected and mapped in both cases and the observations are currently being analyzed. Both galaxies appear to have extensive HI disks extending well beyond the optical rings. It is clear that the next few years will bring new and interesting dynamical information to add to the mainly photometric studies of ring galaxies performed so far.

Another area in which new ground is being broken is in the study of molecules in galaxies. Only in the last few years have searches been made for molecular gas in ring galaxies. The most systematic study made so far is by Horellou et al. (1994), using the 30m diameter IRAM telescope in the North and the 15m SEST telescope in the southern hemisphere. Horellou and collaborators have detected 14 out of 16 ring galaxies in the ¹²CO(1-0)115 GHz (millimeter) line and six were also detected in the ¹²CO(2-1)230 GHz transition. The inferred total mass of H₂ in the rings is, on average, twice the value of a control sample of normal galaxies. A strong correlation was found between the implied H₂ mass and the far IR luminosity of the galaxies, suggesting that the existence of copious molecular material encourages high star formation rates. In contrast, like Silverglade and Krumm before them, the authors conclude that the HI content of ring galaxies is similar to that of normal late-type galaxies.

Unlike nuclear starbursts, ring galaxies contain star formation regions exclusively in the ring. It is therefore of considerable interest to see how the molecular material is distributed. Unfortunately, very few of the rings detected by Horrelou et al., were spatially resolved by the 30m IRAM telescope. Those that were, provide no obvious pattern. In IIHz4, emission was observed from the brightest star forming regions of the ring and very little from the center. On the other hand, the case of the Seyfert ring galaxy NGC 985 is quite different. Here the emission is strongly centered on the "empty" region of the ring, although considerable emission is also seen associated with the peculiar star forming "bar" extending from the Seyfert nucleus. A clear understanding of the distribution of molecules in ring galaxies will have to wait mapping of the galaxies with millimeter aperture synthesis telescopes which will provide much higher spatial resolution.

The two ring galaxies not detected in the CO line by Horellou et al. are interesting. One of them is II Zw 28, the small RK ring shown in Figure 10. The upper limit to the total inferred H₂ mass from II Zw 28 is < 8.5 × 10⁸ M. Interestingly, this galaxy is only barely detected as a faint IRAS source, possibly indicating a lower level of star formation activity than other rings. on the other hand, IIZw 28 has a high H luminosity similar to other ring galaxies observed by Marston and Appleton (1995). II Zw 28 is one of the few galaxies to exhibit unusually strong Balmer absorption lines interior to the ring (Sargent 1977; Charmandaris and Appleton 1994) and it is conceivable that a major burst of star formation has occurred in the recent past which has used up a large fraction of the molecular reservoir in this galaxy, as well as depleting its dust content. New HST observations (Appleton et al., in preparation) show two possible companion galaxies, one of which is of very low surface brightness.

Figure 10. The RK ring galaxy II Zw 28, like Arp 147, was well known as an empty ring with no obvious companion. This I-band Hubble Space Telescope image shows both a possible companion seen just inside the ring. (Appleton et al. in preparation). (See Color Plate VI at the back of this issue.)

The other non-detection in the ¹²CO (J = 1-0) line (made with SEST 43 arcsec beam) is the Cartwheel ring. Horellou et al place an upper limit on the H₂ mass of M(H₂) < 1.5 × 10⁹ M). Because of the large angular size of the Cartwheel, a 5-point map was made and the total on-source integration time was 34 hours. Despite the non-detection of CO from the galaxy (the rms noise was 0.6mK in a 30 km s^-1 channel after 34 hours of integration), the authors admit that the limit is not especially stringent. For example, they point out that the lower limit to the ratio of IRAS FIR flux to the molecular hydrogen mass, L(FIR) / M(H₂), is > 17.7, which is not unusually high. Neither is the upper limit to the ratio of M(H₂) / L_B < 0.024 unusually low, but lies within the range for normal galaxies. Other searches for CO from the Cartwheel have so far been inconclusive (Higdon, personal communication). As Horellou et al. point out, the Cartwheel may well contain significant quantities of molecular hydrogen, but may be deficient in carbon and therefore CO molecules. Oxygen atoms were also found to be deficient in the outer ring HII regions (Fosbury and Hawarden 1977).