Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 325, 933-942 (1997)

Previous Section Next Section Title Page Table of Contents

4. Discussion and conclusions

We have analyzed the data on global structural properties of all known PRGs, and came to a conclusion that there is a correlation between the main characteristics of polar rings and host galaxies. There exist two different classes of PRGs: bulge-dominated galaxies with internal (relative to the optical size of a galaxy) rings and disk-dominated systems with extended rings. The sample studied is somewhat small (8-9 objects), but when we combine our preliminary inference with the results of numerical experiments, a plausible picture of PRG properties emerges.

We have presented the investigation of the ring-forming process by studying the stripping of gas-rich galaxy outskirts during the encounter with galaxies of different structures and the subsequent settling of the captured material in the potential of a companion galaxy using an SPH method. The simulations have demonstrated that the ring sizes depend on galaxy potentials. Under the same interaction conditions, rings twisting around galaxies with a strong concentration of mass to the center (bulge-dominated systems) are less extended (on the average) than that around disk-dominated galaxies. Moreover, we have found that it is impossible to obtain very extended rings (with a diameter of about 30 kpc) taking into account the distribution of the luminous mass only. To obtain such extended rings, the galaxy mass distribution should be less concentrated than observed luminosity distribution. Therefore the presence of dark halos is required to explain the existence of extended rings.

One can suppose that the observational dichotomy is the result of selection. Indeed, interactions between galaxies accompanied by matter accretion occur with participation of various types of galaxies. But we have more chances to create long-lived extended rings when the accreting galaxy is a gas free, disk-like galaxy with extended massive dark halo. So such extended polar rings are expected to be observed preferably around specific types of galaxies.

The existence of the observational dichotomy suggests that PRGs with bulge-dominated central galaxies may possess less pronounced dark haloes in comparison with the first group of objects (another possibility - bulge-dominated galaxies have more centrally-concentrated halos - does not change our results as follow from Sect.3.1.3). These conclusions do not contradict the present-day observational data about dark halos in early-type galaxies (e.g., Bertin & Stiavelli 1993, de Zeew 1995). Nevertheless, we do not exclude other explanations of the revealed observational dichotomy (for instance, environmental influence etc.).

As was mentioned in Sect. 2, there is a close similarity between extended rings and disks of late-type spiral galaxies. Extended rings have the same total luminosities, integral colors, ratios of HI mass to the blue luminosity corrected for internal absorption, gas-to-dust ratios as normal spiral galaxies; there is evidence for the gaseous medium in polar rings to be near (or above) the gravitational stability limit, as the gaseous disks in spiral galaxies (Reshetnikov et al. 1994). Two polar rings (in UGC  7576 and UGC  9796) demonstrate H [FORMULA] -derived global star formation rates to be typical for normal spiral galaxies; general properties of HII regions in these galaxies are normal for late-type spirals (Reshetnikov & Combes 1994a). HII regions in the ring of prototype polar-ring galaxy NGC  2685 demonstrate nearly solar oxygen abundances (Eskridge & Pogge 1994). Extended rings show radial color gradients with bluer colors at larger radii (Reshetnikov et al. 1994, Arnaboldi et al. 1995). The HI/ [FORMULA] mass ratio in the rings of NGC  660 and NGC  4650A is usual for spiral galaxies (Combes et al. 1992, Watson et al. 1994). There is evidence for the spiral structure in the rings of UGC  7576 (Reshetnikov & Combes 1994b) and NGC  4650A (Arnaboldi et al. 1996). There is an indication of the presence of a large-scale magnetic field in the ring of NGC  660 (Reshetnikov & Yakovleva 1991). Moreover, according to Combes & Arnaboldi (1996) and Arnaboldi et al. (1996), dark matter in PRGs could co-exist with the HI component, leaving its mark on the kinematics of the polar ring itself similar to the situation with a late-type galaxy gaseous disk. One can note finally that PRGs lie close to the Tully-Fisher relation for spiral galaxies (Knapp et al. 1985). Therefore, PRGs with extended rings could be undistinguishable from normal spiral galaxies at less advantageous orientation (with a more face-on ring) - see the discussion of this question in PRC. Probably, they will look like early-type spiral galaxies (maybe barred) with extended low surface brightness disks.

The striking similarity between extended rings and spiral disks suggests that PRGs are giant low surface-brightness spiral galaxies with decoupled bulges. Although this cannot be completely ruled out in some special cases, we suppose that the ring capture due to external accretion is a more frequent process (see Sect. 1 about observations of forming rings). Therefore, in contrast to the usual opinion that galactic interactions lead to fast morphological changes towards S0/E, the observational data on PRGs and our modelling simulations provide the example of a rare opposite shift - from an early-type gas-free galaxy through the capture of a gaseous disk to a spiral galaxy. One can suppose that PRG formation due to external accretion can imitate in some aspects the formation of disk galaxies in the hierarchically clustering model of the Universe (e.g., Steinmetz 1996).

In conclusion, let us reply shortly to the main questions of the article (see Sect. 1):

the accretion of matter from a gas-rich companion is a quite effective mechanism for polar-ring formation;

the spatial size of the forming ring is determined (under the same interaction conditions) by the mass distribution of the galaxy;

the process of ring formation takes approximately [FORMULA] years for haloless systems and reaches up to a few Gyrs for host-galaxies possessing massive halos;

the total mass captured into the ring during an encounter with a gas-rich spiral galaxy exceeds [FORMULA].

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: April 28, 1998