Kirk D. Borne
Hughes STX, NASA Goddard Space Flight Center, Code 631, Greenbelt, MD 20771
Ray A. Lucas
Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218
Philip Appleton, Curtis Struck
Iowa State University, Department of Physics & Astronomy, Ames, IA 50010
Alfred B. Schultz
Computer Sciences Corporation, STScI, 3700 San Martin Drive, Baltimore, MD 21218
University of Nevada - Las Vegas, Department of Physics, Las Vegas, NV 89154
Formerly at the STScI. STScI is operated by AURA, Inc. under contract to NASA.
The primary ring around the galaxy is full of blue star-forming knots, which are well resolved in our images. The ``spokes'' of the cartwheel, which are interior to the primary ring, are clearly visible, though their structure is somewhat more amorphous than that of the outer ring. There is a well defined secondary ring around the core and a lens or disk interior to that. This region is heavily reddened and clearly full of dust. The images reveal a very fine structure in the dust distribution around and interior to the secondary ring. A nearly point-like nucleus is seen within the inner disk. The eastern-most companion shows very weak evidence of interaction, shows no evidence for star formation or gas, and appears to be an S0 galaxy. On the other hand, the western-most companion is very disturbed, shows strong evidence for star formation and gas, and appears to be a disrupted late-type spiral or irregular.
Keywords: Interacting Galaxies, Star Formation, Starbursts
Studies of interacting and merging galaxies have intensified in recent years as it has become clear that significant amounts of star formation can take place in collisions of gas-rich systems (e.g., Larson & Tinsley 1978, Lonsdale, Persson, & Matthews 1984, Joseph & Wright 1985, Keel et al. 1985, Bushouse 1986, Kennicutt et al. 1987, Kennicutt 1990, Smith & Kassim 1993, Keel 1993, Smith et al. 1995, Liu & Kennicutt 1995a,b). We have used the high-resolution imaging capabilities of the HST Wide Field and Planetary Camera II (WFPC2; Trauger et al. 1994) to study tidal shocks and star formation within a very special class of such systems: the collisional smoke-ring galaxies.
Of the many types of interacting galaxies known, the rare and beautiful smoke-ring galaxies are among the most straightforward to interpret dynamically. Since the pioneering models of Lynds & Toomre (1976) and Theys & Spiegel (1977) it has become accepted that many of the ``classical'' ring galaxies are formed from a head-on collision between a small intruder galaxy and a larger disk system. The ring forms as gas and stars are crowded into an expanding wave that moves radially through the disk. The passage of the wave triggers vigorous star formation in the rings and provides us with a remarkably simple example of density wave-induced star formation (Jeske 1986, Appleton & Struck-Marcell 1987a,b, Struck-Marcell 1990, Gerber 1993, Struck-Marcell & Higdon 1993, Gerber & Lamb 1994, Mihos & Hernquist 1994). Recent ground-based observations of rings (Marcum, Appleton, & Higdon 1992, Marston & Appleton 1995, Higdon 1995) reveal stellar evolutionary effects in and behind the expanding rings that strongly support the collisional origin.
In order to explore in detail the process of propagating star formation in these remarkable systems, one must probe deeply into the structure of the luminous H II regions that lie on the outer edge of the density wave. Unlike many regions of massive star formation found in astronomy, the site of the youngest (most massive) stars in rings is continually expanding, leaving behind in its wake a trail of evolving stars. The high spatial resolution of HST provides an exciting opportunity to investigate the properties of the massive O/B associations formed in this expanding wave.
Evidence for high star formation rates was obtained in the late 1970's and early 1980's for several ring galaxies. Theys & Spiegel (1976) made the first systematic study of rings and found blue colors in their rings. Several additional studies were made of individual systems which led to the conclusion that vigorous star formation was indeed occurring in the rings (Fosbury & Hawarden 1977, Thompson & Theys 1978, Few, Madore, & Arp 1982), a result later confirmed by IRAS observations (Jeske 1986, Appleton & Struck-Marcell 1987a, Wakamatsu & Nashida 1987). In addition, ground-based optical and IR imaging observations reveal color gradients across and inside some rings (Marcum, Appleton, & Higdon 1992, Appleton, Lysaght, & Marston 1992, Higdon 1993). The large-scale gradients that have been observed so far can be interpreted as evidence that young stars that are born in the expanding ring will evolve and redden in its wake. Other observations indicate that shock waves within the expanding density wave may also play an important role in the star formation process (e.g., Schultz et al. 1991, Lamb & Gerber 1992).
Ground-based imaging has already provided tantalizing evidence for rapid changes in both the stellar and gaseous conditions within ring galaxies on the timescale of the ring propagation (a few times yr). However, the most dramatic changes are likely to occur on the leading edge of the expanding wave where the most massive stars will be found. These H II regions are dominated by stars more massive than 10 M, whose lifetimes are little more than a few million years. If such stars were born on the edge of a ring expanding at typically 100 into the surrounding disk, then the extremely massive O stars would exist in a band only 0.3kpc wide. At the typical distance of a ring galaxy (100Mpc) this corresponds to about 0.6. It is therefore evident that the very high angular resolution of the HST is particularly suited to the investigation of the formation and distribution of the most massive stellar associations within such systems. Furthermore, since the number of nearby ``canonical'' smoke-ring galaxy candidates is quite small (i.e., the nearest examples have redshifts in the range --9000 ), analysis of the crucial questions related to star formation in the expanding ring density wave demands HST resolution. A number of these systems show remarkable fine-scale structure in ground-based images (e.g., the ``Cartwheel'' ring galaxy has a dazzling array of knots and spokes), but most of the details in those structures are simply unresolved from the ground. The real business of density wave-induced star formation is occurring on scales that only HST can begin to resolve.
Multicolor ground-based optical/IR photometry of the Cartwheel shows strong evidence for large B-V and V-K radial color gradients that are consistent with the evolution of a massive stellar population left behind after the passage of the density wave-induced star formation event. Can we observe differences in the colors of individual star clusters in the ring H II regions as one progresses from the outer edge of the ring to the inner edge? Such differences would result from a mix of very high mass (30--60M) O stars being formed along with the more long-lived 10--30M stars. If the stars are born with an IMF that extends to the high masses then the outer edge of the rings should be very luminous and extremely blue resulting in a very sharp outer edge to the rings on angular scales of 0.3--0.5arcsec. On the other hand, if the IMF is truncated at 20M or less then the ``edge'' would be less blue and more extended, since the newly formed clusters would be more long-lived. The unique nature of the propagating ring wave will provide a tool for exploring these changes at HST resolution.
To address the issues discussed here we have obtained HST images of one of the most spectacular examples of collisional ring galaxies from the Arp-Madore (1987) atlas of peculiar galaxies: AM0035-335, commonly referred to as the ``Cartwheel'' galaxy. As evidenced in its complex knotty substructure, the Cartwheel shows clear evidence for strong star formation in its ring, presumably induced by the collision event that formed the ring. The Cartwheel is one of the best examples of well formed ring galaxies with their nucleus still present (i.e., not being disrupted by the recent head-on encounter with the intruder) and yet with some uncertainty as to the present identification of the former intruder (perhaps being one of the small galaxies seen in the surrounding field; but see section 4.9). This galaxy is large (35kpch) and has a ``filled'' ring, showing faint unresolved structures in the form of ``spokes'' within the ring (hence the name ``Cartwheel''). It also shows a second smaller ring around the core.
The Cartwheel ring galaxy (AM0035-335) was observed with the HST WFPC2 on October 16, 1994. We obtained four broad-band B (filter F450W) images totaling 3200 sec exposure time, and four I (F814W) images totaling 1800 sec. The I-band image is useful for mapping the old stellar distribution within the galaxies and around the rings, while the B-band image is useful for tracking the color variations in those regions, thereby tracing the distribution of dust, H II regions, and other young star-forming regions.
For all of the WFPC2 images of the Cartwheel (whose diameter ), the center of the galaxy was placed in WF3 near the boundary of the WF3 and WF4 fields-of-view, thereby allowing the entire galaxy to be studied in the WF3 and WF4 fields and also allowing the two nearby companions (possible collision partners responsible for the ring formation) to be studied in detail in the WF2 field. Both the B and I band images were obtained in sets of two images, with one pair slightly shifted (by 1 east and 2 south) relative to the other pair. This was to compensate for any hot pixels or other CCD blemishes and defects coincident with features in the observed target. The full set of four images in each band was used to remove cosmic ray events from each individual image. The calibration and photometric analysis of the images were carried out using the standard prescription and parameters, as described by Holtzman et al. (1995a,b).
We have combined the B and I band images to form an approximately true-color image of the Cartwheel system. The balance between the B and I band was set by assuming that the eastern-most companion (whose morphology suggests that it is an S0 galaxy) has the B-I color of a typical S0. The full-color rendition of the HST Cartwheel observations have been widely distributed (e.g., see the press release version of the image at the STScI Public Outreach WWW site). We simply show here (in Figure 1a) a grayscale rendition of the B-I color variations within the mosaic'ed WFPC2 image. This figure reveals a complex pattern of sharp color variations in and around the Cartwheel, with the bluest colors found in the outer ring and the reddest colors found in the inner regions. This figure does no justice to the wealth of fine-scale detail revealed in the original images. We show in Figure 1b a grayscale reproduction of one small section of our images (a section from the southern part of the Cartwheel ring), which does show the complex pattern of star-forming knots (possibly young globular clusters or other massive stellar associations) at the highest resolution limits of the WFPC2.
Figure: (a) (left) This image represents the B-I color map of the Cartwheel galaxy as observed with the HST WFPC2. The bluest regions appear as dark shades of grey; the reddest regions appear as the lightest shades. Note the color variations around the ring, within the Cartwheel, and between the two companion galaxies (as described in the text). North is to the upper left; east is to the lower left. (b) (right) An enlargement of the southern section of the Cartwheel ring, showing a small arc of emission extending out from the main ring. A knot is seen in the arc (possibly indicative of star formation induced in the expanding arc).
Figure: (a) (left) A surface plot of the B-band light coming from the central region of the Cartwheel galaxy (showing the central ring and the very sharp nucleus, plus a ragged appearance due to dust and young star-forming regions). (b) (right) A surface plot of the I-band light coming from the central region of the Cartwheel (showing a much smoother appearance than the B-band image).
We show in Figure 2 a pair of surface density plots for the innermost regions of the Cartwheel, derived from the B-band (Fig. 2a) and I-band (Fig. 2b) images. It is clear in comparing the two figures that there is a significantly smoother component to the light in the I band than in the B band. For example, the core region and the inner ring reveal point sources and a more patchy luminosity distribution in the B band. The original full-color image shows that the outer ring is also clearly dotted with a number of blue point sources. Surprisingly, the spokes interior to the outer ring (though somewhat blue) remain amorphous even at high resolution, probably indicative of older expanding stellar associations comprising the spokes.
The two nearby companions also reveal distinctive properties in Figure 1a. The western-most companion is irregular, though similar to a small barred spiral, contains a collection of blue point sources, and has no particular brightness peak at the center. The eastern-most companion appears to be an S0 galaxy, predominantly red in color, with a faint (tidal?) tail of emission extending to the east.
Figure 1a portrays a remarkable richness of information about the Cartwheel galaxy and its companions. Its stunning contrasts are also quite evident: the redness of the S0 companion compared to the Irr (SB?) companion, the red core and red inner ring of the Cartwheel compared to the intense blue outer ring. The star formation on the south side of the outer ring is demonstrably more intense than that on the north side. The spokes are indeed comprised of young stellar populations, though not as intensely blue (i.e., young) as the outer ring. The core also reveals a complex structure not discernible from the ground-based images: dust lanes cutting across and passing around the inner ring, a lens-like ``bulge'' region interior to the inner ring, and a sharp nucleus (Fig. 2).
The various features of the HST images that are identified above will be discussed in the following sections.
The ring annulus is nearly 5 across, thereby comprising many WFPC2 resolution elements across the ring. The ring demonstrates a wealth of sharp features: star-forming knots, stellar clusters, arclets (Fig. 1b), bubble-like regions, and holes in the ring (blast regions?). The arclets and bubbles may be similar to the vast multi-supernova-driven arc reported by Vader & Chaboyer (1995) for NGC 1620.
The spoke region of the Cartwheel is surprisingly amorphous. This would not have been obvious initially from the ground-based images, which showed a complex knotty structure both in the outer ring and in the spokes. The diffuseness of the spoke region is likely a result of the aging and diffusion process for the expanding stellar associations that lie in this region---in the wake of the expanding ring density wave. Numerous circular regions (bubbles?) and young stellar clusters are seen among the spokes, both probably indicative of the earlier violent star formation history within this region.
The HST images reveal a significant level of detail in the core of the Cartwheel (Figs. 2a and 2b). There is a clear lens-like core interior to the inner ring, a sharp nucleus within the core, some bluish clusters around the inner ring, a complex pattern of small-scale dust features, and a three-dimensional appearance (with the spokes continuing across, behind, and into the inner ring). The inner ring is likely illuminated by older stars (from the collision-induced starburst) now ascending the giant branch.
The HST images of the inner regions of the Cartwheel reveal many fine-scale dust features in and around the core. Marcum, Appleton, & Higdon (1992) found very little evidence for a pre-existing stellar population outside the core. The redness of the central region is thus likely a combination of the older stellar population and the significant dust component embedded therein. They also found that most of the HI is concentrated in the outer ring, which is the bluest (and hence most intensely star-forming) region of the galaxy. Unlike the core region, there is probably very little contamination of the young stellar populations in the outer ring and spoke regions with an older population.
One of the goals of the HST observations that has been spectacularly fulfilled has been the detection of color gradients in the ring. Gradients are seen around the ring (bluest on the south side) and across the ring, where there is a different sense to the color gradients on opposite sides of ring (see Fig. 1a). The most intense star formation is on the south side of the outer ring, which is sharper, bluer, and shows more young stellar associations (young globular clusters?). On the south side, the color gradient across the ring is such that the outer part of the ring is bluest; the opposite is seen on the north side of the ring. This peculiar change in the color gradient from south to north has in fact been seen in numerical models by Mihos (private communication)---the result is due to the three-dimensional warping of the ring plane, thereby projecting the ``physically outermost'' (and hence bluest, most intense star-forming) region interior to the ``projected outermost'' region on the north side, and vice versa on the south side. If this interpretation is correct, then the observed color variations are evidence for the expected three-dimensional character of this very disturbed collision remnant.
Many authors have now reported the detection of young globular clusters in HST images of merging galaxies. These clusters are apparently forming in the shocked gas components of the constituent gas-rich galaxies involved in the collisions. For a discussion of these observations and their interpretation, refer to Holtzman et al. (1992), Ashman & Zepf (1992), Zepf & Ashman (1993), Whitmore et al. (1993, 1995), Shaya et al. (1994), Conti & Vacca (1994), Zepf et al. (1994, 1995), and Whitmore & Schweizer (1995). We believe that we are seeing a very rich population of similar objects throughout the outer Cartwheel ring. We hope to measure the properties of these dense star-forming knots, though this analysis will be complicated by the strong crowding of the clusters all around the ring.
The two companions of the Cartwheel that have been included in our HST images are a study in contrasts. As described earlier (section 4.1), we see a smooth, red S0 companion with a tidal tail versus a starbursting non-nucleated Irr companion. It was originally thought that one of these two galaxies was the intruder that produced the Cartwheel ring morphology. But, this interpretation is now in doubt (see section 4.9). It is possible that the disturbed morphologies of these two nearby companions were induced by their own mutual gravitational interaction, perhaps even independent of the presence of the Cartwheel.
Recent HI 21cm imaging observations of the Cartwheel galaxy and its surrounding environment have implicated a third more distant companion as the ``bullet'' that hit the target galaxy that subsequently became the Cartwheel ring galaxy (Higdon 1995). There is an HI tail connecting the Cartwheel with this third companion (to the north). However, the result is not entirely conclusive since the HI tail is seen in only one velocity channel. In any case, it appears certain that small-group (N=4) dynamics involving multiple galaxy-galaxy interactions will need to be invoked in any scenario used to explain both the formation of the Cartwheel ring and the peculiar appearances of its two nearest companions.
The fact that the Cartwheel galaxy is not exactly symmetric indicates that it was formed in an off-center collision with its intruder. The expansion rate of the ring (60 ) and its diameter (35kpch) suggest an age 300 million years. This is consistent with the distant third companion being the intruder and with the two nearest companions interacting independently within the same group (i.e., not causing the ring). Ring galaxy models show that the first ring is often weak and propagates rapidly off of the disk, while a stronger second ring propagates more slowly across the disk. The observed strength of the Cartwheel ring is consistent with it being the second ring, and the apparent age of the collision event is consistent with this interpretation also. Recent results from our HST observations have been reported by Struck, Appleton, Borne, & Lucas (1996) indicating that some of the peculiar features seen near the core region of the Cartwheel are in fact clumps of material raining back down onto the galaxy following the big splash that formed the ring morphology. The expected timescale for this material to fall back into the galaxy is again consistent with the collision age and ring age for the system.
Recent numerical models of ring galaxies (including hydrodynamical terms) show that the formation of ``spokes'' is a natural consequence of instabilities in the gas behind the expanding ring density wave. These ``spokes'' are not, as some have suggested, simply the original spiral arms of the target galaxy reappearing. They are a separate gravitational collapse phenomenon, produced in the shearing, shocked gas behind the expanding wave.
HST has provided the unique opportunity to perform high-resolution imaging of the point-like and sharp-edged features produced in ring galaxies that are formed as a result of their clean head-on collision with a smaller companion galaxy. Such ring galaxies provide an ideal laboratory for studying the gas dynamical response in strongly penetrating galaxy collisions and for studying in particular: the properties (e.g., amplitudes, locations, phase lags) of the propagating wavefronts of star formation and shocked compressed gas, the size and spatial distribution of star-forming regions and H II complexes, and the potential production of globular clusters in the wake of the expanding ring. Most of these features in the galaxies are not resolvable from the ground. In particular, all of the most-structured ring galaxies are at redshifts that place them just beyond the resolution limit of ground-based facilities, with the thickness of the rings being only a few arcseconds across.
Our detail-rich HST images of the Cartwheel have revealed a very fine-scale structure in the morphology of the ring, in its color gradients, and in its dust distribution. The putative collision scenario ascribed to the formation of the ring is consistent with the observations, including the apparent starburst-collision connection. We find a rich population of young stellar clusters in the system (similar to many other published HST studies of interacting galaxies), numerous arcs, superbubbles, and supernovae indicators, all of which are consistent with the previously derived O star population and H luminosity (Marcum, Appleton, & Higdon 1992). There is a distinct pattern in the colors of the various Cartwheel regions, ranging from very blue in the outer ring, to less blue in the spokes region, to red in the secondary inner ring, and finally to very red in the central disk (or lens) core region. There is an extraordinarily large population of young star clusters in the outer ring, apparently formed during the galaxy-galaxy collision event that produced the ring morphology. A very red S0 companion and a very blue Irr companion galaxy are included in our HST images, with both galaxies showing evidence of a recent tidal disturbance. However, recent ground-based HI 21cm imaging observations by Higdon (1995) point to a third more-distant companion to the north as the likely intruder that produced the Cartwheel ring morphology roughly 300 million years ago.
We are using the HST WFPC2 images of the Cartwheel to map the fine-scale brightness distribution and color variations in the star-forming knots, to trace the line-emitting gas, to delineate the density wave, to trace the star formation history, to measure the sizes and distribution of H II regions, and to measure variations in the stellar population mix across and within the rings. These data will then serve as input and strong observational constraints on new stars+gas numerical simulations of ring galaxies. Our planned theoretical analysis will contribute substantially to an understanding of the interplay between stellar and gas dynamical processes in the collision of gas-rich galaxies. In the end, we hope to characterize the small-scale structure observed in the rings in terms of starburst strength, density wave strength, magnitude of gas compression, scale length of shocks, location of star-forming regions, potential for supernovae, and time elapsed since the collision. These goals will be met if we are successful in finding a specific collision model for the system, allowing us to derive plausible physical parameters of the expanding ring (e.g., age, star formation rate, gas distribution, and flow velocity).
The early ring galaxy models of Lynds & Toomre (1976) and Theys & Spiegel (1977) utilized very simple stellar dynamical codes and were quite successful in reproducing the observed ring morphologies. Appleton & Struck-Marcell (1987b) and Struck-Marcell & Appleton (1987) performed detailed fluid-flow calculations, but did not include any of the stellar dynamical effects in their models. Even though gas dynamical processes are very complex and difficult to model, the existence of an approximate symmetry axis during the collision event provides a tremendous leap over many of the complexities. A number of authors (e.g., Lamb & Gerber 1992, Hernquist & Weil 1993, Struck-Marcell & Higdon 1993, Gerber 1994, Lamb, Gerber, & Balsara 1994, Struck 1994, and Mihos & Hernquist 1994) have recently carried out self-consistent stars+gas hybrid simulations of ring galaxy formation, while much work has been done on more general interactions as well (e.g., Olson & Kwan 1990a,b, Barnes & Hernquist 1991, Mihos, Richstone, & Bothun 1991, Bekki & Noguchi 1994, and references therein).
The most recent SPH simulations of rings in gas disks resolve smaller scales previously possible. With the inclusion of self-gravity, they are now able to reveal the formation of ``spokes'' and self-gravitating knots in the rings, as well as strong shocks. The ring galaxy models of Struck-Marcell & Higdon (1993) are especially relevant to this study. They not only provide detailed simulations of the Cartwheel ring, but they also provide an example of how other systems could be modeled. However, most of these SPH simulations used an isothermal equation of state. Heating, cooling, and the thermal feedback effects of young star activity are expected to play an important role in determining small-scale structure in the rings, which has motivated us to develop more physically realistic (but simplified) heating and cooling functions for our SPH codes. This is where HST observations will play a crucial role, revealing the structures on the smallest scales (e.g., shock fronts, H II regions, star-forming complexes, stellar associations, globular clusters). On the global scale, we hope to compare the dynamical (i.e., collision) age of the ring with the apparent age of the young stellar population in order to examine the connection between the star formation history and the passage of the ring density wave through the galactic disk. On smaller scales, we will test the predictions of our recent models as well as those of our earlier models (Appleton & Struck-Marcell 1987b, Struck-Marcell & Appleton 1987); the latter predict a definite phase lag between the peak of the density wave and the peak star formation rate. All of these various properties can be modeled with the new numerical simulation algorithms. It is hoped that such models, when matched to our detailed HST imaging observations, will ultimately elucidate the prevailing physical (stellar and gas dynamical) processes and conditions that apply within collision-induced ring galaxies such as the Cartwheel.
Support for this work was provided by NASA through grant number GO-5410.01-93A from the Space Telescope Science Institute, which is operated by AURA, Inc., under NASA contract NAS5-26555. We thank Brad Whitmore and members of the STScI WFPC2 Analysis Team for many helpful discussions and for providing the aperture photometry script used for portions of the analysis reported here. We also thank the Director of the STScI for the allocation of observing time on the HST. We especially thank Louis ``Eddie'' Bergeron for valuable assistance in the analysis and manipulation of the images.
Appleton, P. N., Lysaght, M., & Marston, A. P. 1992, BAAS, 23, 1391
Appleton, P. & Struck-Marcell, C. 1987a, ApJ, 312, 566
Appleton, P. & Struck-Marcell, C. 1987b, ApJ, 318, 103
Arp, H. C. & Madore, B. F. 1987, A Catalogue of Southern Peculiar Galaxies and Associations, (Cambridge: Cambridge University Press)
Ashman, K. M. & Zepf, S. E. 1992, ApJ, 384, 50
Barnes, J. E. & Hernquist, L. E. 1991, ApJ, 370, L65
Bekki, K. & Noguchi, M. 1994, A&A, 290, 7
Bushouse, H. 1986, AJ, 91, 255
Conti, P. S. & Vacca, W. D. 1994, ApJ, 423, L97
Few, J. M. A., Madore, B. F., & Arp, H. C. 1982, MNRAS, 199, 633
Fosbury, R. A. E. & Hawarden, T. G. 1976, MNRAS, 178, 473
Gerber, R. A. 1993, PhD Thesis, University of Illinois
Gerber, R. A. 1994, BAAS, 26, 1430
Gerber, R. A. & Lamb, S. A. 1994, ApJ, 431, 604
Hernquist, L. & Weil, M. L. 1993, MNRAS, 261, 804
Higdon, J. 1993, PhD Thesis, University of Texas
Higdon, J. L. 1995, ApJ, 455, 524
Holtzman, J. A. et al. 1992, AJ, 103, 691
Holtzman, J. A. et al. 1995a, PASP, 107, 156
Holtzman, J. A. et al. 1995b, PASP, 107, 1065
Jeske, N. 1986, PhD Thesis, University of California, Berkeley
Joseph, R. & Wright, G. 1985, MNRAS, 214, 87
Keel, W. C. 1993, AJ, 106, 1771
Keel, W. C., Kennicutt, R. C., Hummel, E., & van der Hulst, J. M. 1985, AJ, 90, 708
Kennicutt, R. C. 1990, in Paired and Interacting Galaxies, IAU Colloquium #124, eds. J. W. Sulentic, W. C. Keel, & C. M. Telesco (NASA), 269
Kennicutt, R. C., Keel, W. C., van der Hulst, J. M., Hummel, E., & Roettiger, K. A. 1987, AJ, 93, 1011
Lamb, S. A. & Gerber, R. A. 1992, in The Evolution of Galaxies and their Environment, proceedings of the 3rd Teton Conference (NASA), 225
Lamb, S. A., Gerber, R. A., & Balsara, D. S. 1994, BAAS, 26, 1430
Larson, R. B. & Tinsley, B. M. 1978, ApJ, 219, 46
Liu, C. T. & Kennicutt, R. C. 1995a, ApJ, 450, 547
Liu, C. T. & Kennicutt, R. C. 1995b, ApJS, 100, 325
Lonsdale, C. J., Persson, S. E., & Matthews, K. 1984, ApJ, 287, 95
Lynds, R. & Toomre, A. 1976, ApJ, 209, 382
Marcum, P. M., Appleton, P. N., & Higdon, J. L. 1992, ApJ, 399, 57
Marston, A. P. & Appleton, P. N. 1995, AJ, 109, 1002
Mihos, J. C. & Hernquist, L. 1994, ApJ, 437, 611
Mihos, J. C., Richstone, D. O., & Bothun G. D. 1991, ApJ, 377, 72
Olson, K. M. & Kwan, J. 1990a, ApJ, 349, 480
Olson, K. M. & Kwan, J. 1990b, ApJ, 361, 426
Schultz, A. et al. 1991, BAAS, 23, 953
Shaya, E. J. et al. 1994, AJ, 107, 1675
Smith, D. A. et al. 1995, ApJ, 439, 623
Smith, E. P. & Kassim, N. E. 1993, AJ, 105, 46
Struck, C. 1994, BAAS, 26, 1431
Struck, C., Appleton, P. N., Borne, K. D., & Lucas, R. A. 1996, Nature, submitted
Struck-Marcell, C. 1990, AJ, 99, 71
Struck-Marcell, C. & Appleton, P. N. 1987, ApJ, 323, 480
Struck-Marcell, C. & Higdon, J. L. 1993, ApJ, 411, 108
Theys, J. C. & Spiegel, E. A. 1976, ApJ, 208, 650
Theys, J. C. & Spiegel, E. A. 1977, ApJ, 212, 616
Thompson, L. A. & Theys, J. C. 1978, ApJ, 224, 796
Trauger, J. T. et al. 1994, ApJ, 435, L3
Vader, J. P. & Chaboyer, B. 1995, ApJ, 445, 691
Wakamatsu, K. & Nashida, M. T. 1987, ApJ, 315, L23
Whitmore, B. C., Schweizer, F., Leitherer, C., Borne, K., & Robert, C. 1993, AJ, 106, 1354
Whitmore, B. C. & Schweizer, F. 1995, AJ, 109, 960
Whitmore, B. C. et al. 1995, ApJ, 454, 73
Zepf, S. E. & Ashman, K. 1993, MNRAS, 264, 611
Zepf, S. E. et al. 1994, BAAS, 26, 1431
Zepf, S. E., Carter, D., Sharples, R. M., & Ashman, K. 1995, ApJ, 445, L19