Super star clusters are young, compact star clusters found in the central regions of interacting galaxies. Recently, they have also been reported to preferentially form in certain tidal tails, but not in others. In this paper, we have used 21 cm HI m
aps and the Hubble Space Telescope Wide Field Planetary Camera 2 images of eight tidal tail regions of four merging galaxy pairs to compare the kiloparsec scale HI distribution with the location of super star clusters found from the optical images. For most of the tails, we find that there is an increase in super star cluster density with increasing projected HI column density, such that the star cluster density is highest when log N(HI) >= 20.6 cm^{-2}, but equal to the background count rate at lower HI column density. However, for two tails (NGC 4038/39 Pos A and NGC 3921), there is no significant star cluster population despite the presence of gas at high column density. This implies that the N(HI) threshold is a necessary but not sufficient condition for cluster formation. Gas volume density is likely to provide a more direct criterion for cluster formation, and other factors such as gas pressure or strength of encounter may also have an influence. Comparison of HI thresholds needed for formation of different types of stellar structures await higher resolution HI and optical observations of larger numbers of interacting galaxies.
We present deep HST ACS images of the post-starburt shell galaxy AM 0139-655. We find evidence for the presence of three distinct globular cluster subpopulations associated with this galaxy: a centrally concentrated young population (~ 0.4 Gyr), an i
ntermediate age population (~ 1 Gyr) and an old, metal-poor population similar to that seen around normal galaxies. The g-I color distribution of the clusters is bimodal with peaks at 0.85 and 1.35. The redder peak at g-I=1.35 is consistent with the predicted color for an old metal-poor population. The clusters associated with the peak at g-I=0.85 are centrally concentrated and interpreted as a younger and more metal-rich population. We suggest that these clusters have an age of ~ 0.4 Gyr and solar metallicity based on a comparison with population synthesis models. The luminosity function of these blue clusters is well represented by a power law. Interestingly, the brightest shell associated with the galaxy harbors some of the youngest clusters observed. This seems to indicate that the same merger event was responsible for the formation of both the shells and the young clusters. The red part of the color distribution contains several very bright clusters, which are not expected for an old, metal-poor population. Furthermore, the luminosity function of the red GCs cannot be fit well by either a single gaussian or a single power law. A composite (gaussian + power law) fit to the LF of the red clusters yields both a low rms and very plausible properties for an old population plus an intermediate-age population of GCs. Hence, we suggest that the red clusters in AM 0139-655 consist of two distinct GC subpopulations, one being an old, metal-poor population as seen in normal galaxies and one having formed during a recent dissipative galaxy merger.
