An HI Survey of Actively Star-Forming Dwarf Galaxies

The Astronomical Journal, 124:191–212, 2002 July # 2002. The American Astronomical Society. All rights reserved. Printed in U.S.A. AN H i SURVEY OF ACTIVELY STAR-FORMING DWARF GALAXIES John J. Salzer1 Astronomy Department, Wesleyan University, Middletown, CT 06459; [email protected] Jessica L. Rosenberg2 Department of Astronomy, University of Massachusetts, Amherst, MA 01003-9305; [email protected] Eric W. Weisstein3 Wolfram Research, Inc., 100 Trade Center Drive, Champaign, IL 61820-7237; [email protected] Joseph M. Mazzarella1 Infrared Processing and Analysis Center, MS 100-22, California Institute of Technology, 770 South Wilson Avenue, Pasadena, CA 91125; [email protected] and Gregory D. Bothun1 Department of Physics, University of Oregon, 1371 East 13th Avenue, Eugene, OR 97403; [email protected] Received 2000 November 24; accepted 2002 March 26 ABSTRACT We present the results of H i 21 cm observations of 139 actively star-forming dwarf galaxies obtained with the 305 m radio telescope at Arecibo Observatory. Our sample consists of all objects cataloged in objectiveprism surveys for UV-excess or emission-line galaxies published prior to the start of the survey that have luminosities below MB =
17.0 and that are located within the declination limits of the Arecibo telescope. Galaxies from the Markarian, Michigan, Case, Wasilewski, Haro, and Zwicky lists are included. The sample spans a wide range of both H i gas content and star formation levels. A total of 122 objects (88%) were detected; 82 galaxies have been observed for the first time in H i. The median velocity width for our sample is 88 km s
1, and the median H i gas mass is 3.0  108 M. In general, the sample galaxies are gas-rich, with an average MH i/LB = 1.3 after correcting for the luminosity enhancement due to the starburst. The progenitors of many of the star-forming dwarfs have higher MH i/LB than typically seen in samples of nearby ‘‘ normal ’’ galaxies, emphasizing their distinct nature. Key words: galaxies: compact — galaxies: dwarf — galaxies: ISM — galaxies: starburst all evidence of any underlying older population of stars. The evolution of these low-mass galaxies remains somewhat mysterious. Beginning with the pioneering work of Searle & Sargent (1972), most studies of these objects have concluded that the star formation event is only short-lived. This is due to the fact that the observed current star formation rates would use up the entire supply of gas in the galaxy on a timescale much shorter than the age of the universe. Thus, actively star-forming dwarf galaxies cannot remain forever in their bursting mode and are destined to evolve photometrically (and perhaps morphologically) into some other kind of dwarf galaxy. It is possible that this brightening and fading could be episodic in nature (see, e.g., Gerola, Seiden, & Schulman 1980). However, it is not at all clear that the observed local population of dwarf galaxies is consistent with this simple model. Where are the post-starburst dwarfs, and what do they look like? Part of the uncertainty in the evolutionary status of these galaxies is due to the relatively small numbers of true dwarf starbursting galaxies that have been studied to date. A major motivation of the present study was to substantially enlarge this sample. H i data for star-forming dwarf galaxies are important, since they provide a measure of the amount of fuel available for driving the currently observed starburst. When used in conjunction with the star formation rates inferred from optical spectroscopy or narrowband imaging, this places 1. INTRODUCTION Actively star-forming dwarf galaxies—also referred to in the literature as H ii galaxies and blue compact dwarfs (BCDs)—represent the subset of low-luminosity galaxies undergoing a strong episode of star formation at the current time. The majority of such galaxies known have been discovered in objective-prism surveys using Schmidt telescopes (e.g., Markarian 1967 and Markarian, Lipovetsky, & Stepanian 1981; MacAlpine, Smith, & Lewis 1977 and MacAlpine & Williams 1981 [Michigan survey]; Wasilewski 1983; Pesch & Sanduleak 1983 and Pesch, Stephenson, & MacConnell 1995 [Case survey]). They usually exhibit a compact, irregular morphology and display intense, narrow emission lines emanating from one or more giant starforming regions. The energy emitted by these ‘‘ starbursts ’’ can, in extreme cases, dominate the light at optical and infrared wavelengths, sometimes to the extent of masking 1 Visiting Astronomer, Arecibo Observatory, which is part of the National Astronomy and Ionosphere Center and is operated by Cornell University under a management agreement with the National Science Foundation. 2 Current address: Center for Astrophysics and Space Astronomy, University of Colorado, Boulder, CO 80309-0389. 3 Participant in the Arecibo Observatory Summer Research Experiences for Undergraduates program. 191 192 SALZER ET AL. constraints on models of the star formation history of these objects. In addition, the velocity widths of the H i profiles give an estimate of the total mass of these systems. Both of these quantities are important for understanding the formation and evolution of these dwarf stellar systems. H i data are particularly important for parameterizing the star formation characteristics of dwarf galaxies, because of the weakness (or total absence) of CO and far-infrared flux in these objects (Salzer & MacAlpine 1988; Schombert et al. 1990; Sage et al. 1992; Taylor, Kobulnicky, & Skillman 1998; Barone et al. 2000). Previous 21 cm observations of objects of this type have generally shown them to be gas-rich, with H i masses often comparable to the mass of their stellar component (Bottinelli et al. 1973; Gordon & Gottesman 1981; Thuan & Martin 1981; Hoffman et al. 1989; Staveley-Smith, Davies, & Kinman 1992; Thuan et al. 1999; Smoker et al. 2000). Roughly two dozen have been observed interferometrically (e.g., Lequeux & Viallefond 1980; Viallefond & Thuan 1983; Viallefond, Lequeux, & Comte 1987; Brinks & Klein 1988; Taylor, Brinks, & Skillman 1993; Taylor et al. 1995; van Zee, Skillman, & Salzer 1998a; van Zee et al. 1998b; van Zee, Salzer, & Skillman 2001; Meurer, Staveley-Smith, & Killeen 1998). These observations show that the H i gas is typically much more extended (by factors of 2–6) than the optical portion of the galaxy and is often lumpy in nature. For example, high-resolution VLA observations of I Zw 18 reveal a very clumpy distribution of H i gas that is embedded in a smoother, very extended envelope (van Zee et al. 1998b). In many of the dwarfs, portions of the H i gas depart from an otherwise orderly velocity field, suggesting turbulence, gas infall, or gravitational perturbations from neighboring objects. In the extreme case, some star-forming dwarf galaxies may in fact be optical cores that are embedded in an extensive and massive H i envelope. At least one low-mass galaxy, NGC 2915 (Meurer et al. 1996), is known to exhibit this characteristic, and we are interested in determining whether similar extreme galaxies might exist in our relatively large sample. Although the previous single-dish 21 cm work carried out was important for establishing the properties of the neutral H i component of these dwarfs, there are a number of reasons for obtaining new, more detailed observations. First, the early studies included objects from only a few surveys (Markarian, Haro, and Zwicky). Since that time, several significantly deeper objective-prism surveys have been undertaken (e.g., Michigan, Wasilewski, Case), which have greatly increased the number of objects available for study. When we began this project, less than 20% of our complete sample of low-luminosity dwarfs had been observed in previous H i studies. Second, most of these previous observations were made in the late 1970s. Substantial improvements in instrumental sensitivity now yield significantly better data than were obtained in the past. This added sensitivity is necessary for probing the low H i masses present in some of our sample galaxies. In x 2, we describe our sample of low-luminosity starforming galaxies and detail the 21 cm observations obtained. Section 3 presents the results of these observations and uses them to develop an understanding of the neutral gas properties of this class of object. In addition, some implications of this study for our understanding of starforming dwarf galaxies are discussed. Vol. 124 2. OBSERVATIONS 2.1. Sample Selection Our sample consists of all galaxies that were cataloged in one of several objective-prism surveys for UV-excess and/ or emission-line galaxies (ELGs) that were known to have B-band absolute magnitudes fainter than
17.0 and were observable from Arecibo. The latter criterion restricts us to galaxies with declinations between
1=5 and 38=5. Objective-prism surveys considered included those by Markarian and Wasilewski, plus the Michigan (UM) and Case surveys. In addition, a few galaxies in the Zwicky (1971) list of compact galaxies and the Haro (1956) list of blue galaxies that satisfied the above criteria were included. Only galaxies whose redshifts were known in late 1988 were included in our sample. The redshift data come primarily from the literature (Gordon & Gottesman 1981; Thuan & Martin 1981; Salzer, MacAlpine, & Boroson 1989a, 1989b; Bothun et al. 1989; Mazzarella & Balzano 1986; Salzer et al. 1995). The declination limits of the Arecibo telescope prevented us from considering sources from a number of other important objective-prism surveys and also forced us to exclude a number of well-known dwarf star-forming galaxies. Table 1 shows the distribution of our sample galaxies from among the various objective-prism surveys used. Listed for each survey are the number of sources observed (Nobs), as well as the number detected (Ndet). In addition, the apparent magnitude and velocity ranges for the detected galaxies from each survey are listed. Note that the sum of the numbers in each of columns (2) and (3) exceeds the totals given in the bottom row; many of the galaxies in our sample are cataloged in more than one of the surveys. More details about the properties of the sample are presented below. The absolute magnitude limit of our survey is fairly arbitrary and represents no particular physical limit. It was chosen to provide a sample of galaxies that truly were dwarfs (i.e., LB < 0.1L*) but at the same time to yield a meaningful sample size. Subsequent to the completion of the H i observations, revised redshifts or improved apparent magnitudes for some of the galaxies in our sample revealed that several sources have MB <
17.0. Since this revelation occurred after the completion of the H i observations, we have decided to retain them in our sample, since in all cases the galaxies are only slightly brighter than the MB =
17.0 limit. It is relevant to compare briefly the characteristics of our sample with those of some previous studies of the H i content of star-forming dwarf galaxies. Gordon & Gottesman (1981) focused primarily on objects from the Haro and TABLE 1 Summary of H i Survey Constituents Survey Name (1) Nobs (2) Ndet (3) mB Range (4) Velocity Range (km s
1) (5) Markarian (Mrk) ....... Michigan (UM).......... Case (CG) .................. Wasilewski (Was)....... Haro .......................... Zwicky....................... Total ...................... 50 39 34 17 9 3 139 43 33 30 17 9 3 122 13.50–17.00 14.10–18.51 13.80–19.07 13.30–17.60 13.80–15.58 14.40–15.55 13.50–19.07 613–3571 882–6912 613–9219 542–5391 648–2024 796–1253 542–9219 No. 1, 2002 H i IN STAR-FORMING DWARF GALAXIES Zwicky compact lists, both of which include a large proportion of nondwarf galaxies. The median absolute magnitude of their sample is roughly
19, or more than a factor of 10 times more luminous than the median of our galaxies. Likewise, the median H i mass is about 10 times higher. Thus, while the Gordon & Gottesman study does include some dwarf galaxies, it is primarily a sample of large galaxies and hence is less relevant for comparison here. On the other hand, the Thuan & Martin (1981) study targeted dwarf galaxies and drew its sample from the Markarian, Zwicky, and Haro lists. With a median MB =
17.3 and a median d25 = 4.5 kpc, this sample is quite comparable to the current list, although the galaxies are slightly larger and more luminous on average. The Staveley-Smith et al. (1992) study of dwarf galaxies includes subsamples of both star-forming and quiescent objects. The optical properties of their sample match closely those in the present study. Finally, the recent paper by Smoker et al. (2000) includes galaxies taken primarily from the UM survey lists and hence has a substantial overlap with our sample. Since Smoker et al. used the fully steerable Lovell Telescope at Jodrell Bank, they could target a number of UM ELGs that lie outside of the Arecibo declination range. Their sample contains a mix of dwarf starforming galaxies and more luminous objects. 2.2. Observing Strategy The data acquisition procedure and instrument setup used for each galaxy were tailored to meet a number of different goals set for the project. We wished to obtain high signal-to-noise ratio H i profiles and accurate fluxes for our entire sample. However, since our sample consisted of dwarf galaxies, we expected that the velocity widths would typically be less than 100 km s
1 FWHM, and in many cases less than 50 km s
1, thereby calling for higher velocity resolution than is often used. In addition, we wished to be sensitive to the presence of profile asymmetries and multiple velocity components. Such asymmetries indicate the existence of substructure in the H i component (e.g., infalling H i clouds, multiple dense H i knots). Galaxies displaying such features would be good candidates for future H i mapping observations; identifying such a sample was an important priority. Therefore, we decided to observe with sufficient velocity resolution to ensure adequate sampling of the profile. Since in most cases we did not know the H i flux level ahead of time, we split the program into two portions. The first consisted of low-resolution observations (20 MHz bandwidth) to determine the peak H i flux level. Each source was observed for a single on-off pair at a frequency corresponding to the optical redshift (which was known in all cases). Depending on the strength (and to a lesser degree the width) of the H i profile, subsequent observations were carried out with 5 MHz (strong sources) or 10 MHz (weaker sources) bandwidths, corresponding to 2 and 4 km s
1 per channel, respectively. Objects that were very weak or undetected in the single 20 MHz scan were usually reobserved at 20 MHz only. Hence, the data presented below have three different velocity resolutions. 2.3. 21 cm Observations Observations were carried out with the 305 m telescope of the Arecibo Observatory on several dates in 1989 May– August and over a 12 day run in 1991 February. The standard 21 cm observing setup and data reduction programs 193 were used, the details of which can be found in Haynes & Giovanelli (1984). We utilized both the 21 cm and 22 cm dual circular feeds. The 22 cm feed was held in a focusing mount that allows coverage of the frequency range 1300– 1395 MHz. It used a HEMT receiver and provided an overall system temperature of 33 K at the zenith. The 21 cm feed, which employed cooled, dual-channel GaAsFET receivers, could not be tuned in frequency in real time, but it could be repositioned in its mount to allow for optimum sensitivity at differing frequencies. During the course of this project, the 21 cm feed was centered at 1418 MHz (1989) and 1414 MHz (1991). The 2048-channel autocorrelator was used in a mode that provides two 512-channel subcorrelators for each of the two independent, oppositely polarized signals. The bandwidth of each subcorrelator was set at either 5, 10, or 20 MHz, the value depending primarily on the source strength (see above). The corresponding velocity resolutions (after smoothing) are 6, 12, and 24 km s
1, respectively. All observations were made in total power mode, alternating every 5 minutes of on-source integration with a 5 minute off-source observation that tracked the same path across the reflector. The number of on-off pairs obtained of each source varied depending on signal strength. Our goal was to obtain high signal-to-noise ratio profiles (S/N > 10 whenever possible) for all objects. The absolute flux density scale and the pointing accuracy of the telescope (about 2000 ) were determined by calibrating the system with strong continuum sources from the catalog of Bridle et al. (1972). The final reductions were made using the GALPAC data reduction package, developed by R. Giovanelli. The 21 cm feed was used primarily for low-redshift galaxies (v  2000 km s
1). In addition, a number of the southernmost sources could only be observed with the 21 cm feed, because of its location on the outside of the feed carriage house (the 21 cm feed had a coverage 1=5 larger in zenith distance than the 22 cm feed). All observations reported here were carried out prior to the recently completed upgrade to the Arecibo telescope. The old 21 cm line feeds have all been retired and replaced by a Gregorian subreflector system and horn feeds. Table 2 lists all 139 dwarf galaxies that were observed at Arecibo, along with relevant optical data. Galaxies are listed in order of increasing right ascension, with objects from the various surveys intermixed. If a given galaxy is included in more than one objective-prism survey (e.g., UM and Markarian or Case and Wasilewski), the alternate designation is given in the last column of the table. The sources for the optical data in the table are given in x 2.5. The contents of Table 2 are as follows: 1. Galaxy name from objective-prism survey; Mrk = Markarian survey, UM = Michigan Survey, CG = Case survey, Was = Wasilewski survey, Haro = Haro survey, Zw = Zwicky compact galaxy. 2. Right ascension (equinox J2000). 3. Declination (equinox J2000). 4. Apparent magnitude in the B band. 5. Galactic absorption in the B band, taken from Burstein & Heiles (1984) if the galaxy is also in the Uppsala General Catalogue (UGC; Nilson 1973). If not, the average AB for the closest UGC galaxies (all those within 1 ) is used. 6. Major-axis diameter measured at the B = 25.0 mag arcsec
2 level, in arcseconds. TABLE 2 H i Survey Sample and Optical Characteristics Name (1) R.A. (J2000) (2) Decl. (J2000) (3) B (4) AB (5) a25 (arcsec) (6) b25 (arcsec) (7) Refs. (8) V (km s
1) (9) Dist. (Mpc) (10) UM 241......... UM 38 .......... UM 40 .......... UM 51 .......... UM 69 .......... UM 80 .......... UM 92 .......... Mrk 1153 ...... UM 102......... UM 323......... UM 330......... UM 112......... UM 336......... UM 345......... UM 133......... UM 372......... UM 382......... UM 396......... UM 408......... UM 417......... Mrk 1038 ...... Mrk 370 ........ 00 25 19.9 00 27 51.5 00 28 26.9 00 37 00.9 00 48 23.6 01 00 21.8 01 18 34.5 01 23 15.3 01 24 40.7 01 26 46.5 01 30 03.5 01 32 39.8 01 33 44.6 01 37 11.1 01 44 41.2 01 50 10.8 01 58 09.4 02 07 26.4 02 11 23.1 02 19 30.3 02 26 28.3 02 40 28.9 00 31 32 03 29 23 05 00 14 03 25 57 04 05 31 04 39 54 02 27 14 01 13 19 03 51 25
00 38 42 00 44 36 04 38 35 00 57 12 00 52 55 04 53 25 02 18 29
00 06 38 02 56 55 02 20 31
00 59 12 01 09 38 19 17 50 17.00 15.50 15.48 17.00 15.40 17.00 18.51 15.70 16.50 16.08 18.07 17.00 18.15 17.91 15.87 15.16 18.56 18.32 17.56 18.20 14.70 13.50 0.05 0.05 0.05 0.04 0.06 0.07 0.06 0.05 0.05 0.12 0.04 0.05 0.04 0.05 0.07 0.04 0.03 0.05 0.04 0.06 0.07 0.25 16 32 37 56 53 16 12 57 8 24 13 57 8 11 64 56 12 8 13 15 66 70 16 24 22 8 22 16 7 13 8 16 9 44 8 8 16 36 7 8 10 7 57 48 6, 3 6, 3 2, 1 6, 3 4, 2 6, 3 1, 1 8, 2 6, 3 2, 1 1, 1 6, 2 1, 3 1, 1 2, 3 1, 1 1, 1 1, 1 2, 1 1, 1 8, 2 8, 2 4238 1377 1343 4198 1960 4784 6912 1769 2188 1914 5093 2016 5759 5672 1622 1700 3747 6206 3492 2584 1499 794 58.3 20.3 19.9 57.8 27.9 65.5 93.5 24.9 30.6 26.7 69.1 28.2 77.9 76.7 22.9 23.7 50.8 83.6 47.4 35.0 20.6 11.9 Mrk 600 ........ II Zw 40 ........ CG 4 ............. CG 6 ............. CG 10 ........... CG 13 ........... CG 14 ........... CG 23 ........... CG 30 ........... CG 31 ........... Mrk 408 ........ CG 34 ........... Haro 22......... Wa 4 ............. 02 51 04.4 05 55 42.6 08 55 33.2 09 00 13.4 09 12 06.6 09 15 27.8 09 15 32.5 09 36 07.7 09 42 57.2 09 43 00.8 09 48 05.1 09 49 55.3 09 50 01.1 09 56 19.6 04 27 09 03 23 31 31 12 42 31 59 57 31 40 51 31 23 17 30 08 35 29 07 02 30 10 21 31 58 51 32 52 56 30 03 46 28 00 49 27 13 55 15.43 15.55 15.58 15.15 17.56 15.86 15.60 15.40 17.67 14.70 14.80 18.13 15.30 14.30 0.12 1.94 0.05 0.01 0.02 0.03 0.02 0.02 0.01 0.02 0.00 0.03 0.02 0.04 38 37 25 31 17 88 32 88 13 32 32 10 40 31 22 17 18 18 10 21 24 16 6 24 24 6 24 26 2, 1 2, 1 3, 1 3, 1 3, 1 3, 1 4, 3 4, 3 3, 1 4, 3 8, 3 3, 1 4, 3 4, 2 1012 796 2009 1956 1928 1876 1890 1640 7621 1645 1559 5188 1455 1239 13.9 9.0 26.1 25.4 25.0 24.3 24.4 21.0 100.8 21.3 20.2 68.4 18.5 15.6 Mrk 411 ........ CG 46 ........... Mrk 714 ........ CG 50 ........... CG 55 ........... Wa 5 ............. Wa 6 ............. Wa 8 ............. Wa 13............ 09 57 32.6 09 59 04.8 10 04 08.7 10 06 18.6 10 08 58.2 10 10 33.3 10 10 58.5 10 19 01.0 10 32 18.0 33 37 03 30 28 31 06 30 38 28 56 45 32 00 58 22 00 42 25 30 30 21 16 56 27 40 14 14.80 18.38 15.20 14.60 16.65 17.50 17.60 15.30 13.30 0.02 0.07 0.03 0.06 0.06 0.07 0.11 0.08 0.05 32 6 32 32 27 17 8 40 93 24 6 24 24 14 13 8 16 48 8, 3 3, 1 8, 3 4, 3 3, 1 5, 1 5, 3 4, 3 4, 2 1463 6266 1260 1371 1441 1290 5391 1083 542 19.0 82.8 14.6 17.5 18.6 15.9 70.8 13.1 6.4 Mrk 724 ........ Mrk 416 ........ Mrk 1263 ...... Mrk 1264 ...... Mrk 1271 ...... CG 75 ........... CG 76 ........... CG 80 ........... 10 41 09.6 10 43 06.1 10 48 56.8 10 49 07.5 10 56 09.1 10 55 48.7 10 56 45.2 11 03 43.3 21 21 43 20 25 08 12 11 42 06 55 04 06 10 22 31 16 28 31 05 33 28 53 07 16.50 15.10 15.40 14.40 14.80 16.50 17.00 13.80 0.01 0.02 0.06 0.04 0.07 0.03 0.01 0.02 16 70 56 40 32 17 14 185 16 22 24 18 24 17 10 40 8, 3 8, 2 8, 3 8, 2 8, 3 3, 1 3, 1 4, 2 1205 1318 1321 712 1044 1265 1631 706 14.9 16.3 15.9 7.4 11.9 16.4 21.2 8.8 CG 82 ........... Mrk 1283 ...... UM 422......... CG 101.......... CG 103.......... 11 04 58.3 11 07 51.5 11 20 14.5 11 20 41.4 11 23 07.0 29 08 24 28 30 03 02 31 51 31 11 03 30 28 49 15.58 15.60 14.10 15.79 15.20 0.00 0.00 0.10 0.00 0.00 33 32 149 18 32 27 24 56 13 24 2, 1 8, 3 1, 1 3, 1 4, 3 648 809 1587 2176 1617 8.0 10.2 19.0 28.6 21.1 194 Other Names (11) IC 52, UGC 494 UGC 931 UGC 1105 UGC 1297 IC 225, UGC 1907 NGC 1036, IC 1828, UGC 2160 UGC 4878 IC 2520, UGC 5335, Re 98 IC 2524 Haro 23 Re 277 NGC 3274, UGC 5721, Re 481 UGC 5833 UGC 5923 NGC 3510, Haro 26, UGC 6126 Mrk 36, Haro 4 UGC 6345B, VV 795 Re 1087, PB 2610 TABLE 2—Continued B (4) AB (5) a25 (arcsec) (6) b25 (arcsec) (7) Refs. (8) V (km s
1) (9) Dist. (Mpc) (10) 20 34 50 15.00 0.00 62 31 4, 2 1410 17.8 11 30 27.2 11 35 23.8 11 35 26.7 11 36 36.7 11 39 00.7 11 40 48.2 11 41 07.4 11 41 30.4 11 41 36.3 11 41 50.5 11 43 21.9 11 43 32.9 11 45 37.0 11 46 26.1 11 47 00.6 11 49 30.2 11 50 02.7 11 50 23.7 11 50 36.2 11 51 33.0 11 52 27.7 11 52 37.1 11 52 47.5 11 54 12.2 36 44 07 31 39 30 33 18 12 00 49 01 31 29 10 35 12 31 32 25 39 32 20 51 32 16 57 15 58 24 31 27 09 31 27 32 31 18 14 34 51 07
00 17 38 24 08 49 15 01 24
00 31 41
00 34 04
02 22 22 34 53 41
02 28 08
00 40 08 00 08 15 15.20 16.55 15.50 15.14 16.32 14.90 15.50 15.50 16.70 14.50 15.50 15.50 15.00 15.00 15.55 16.90 15.49 17.10 15.51 16.27 16.50 14.70 17.87 14.14 0.00 0.00 0.00 0.04 0.00 0.01 0.00 0.00 0.00 0.04 0.03 0.03 0.04 0.00 0.04 0.06 0.09 0.04 0.04 0.04 0.00 0.04 0.04 0.03 48 13 48 39 19 48 32 32 40 22 44 32 40 32 40 16 27 17 45 20 24 31 14 23 24 11 9 24 14 24 16 24 16 13 31 16 24 24 22 16 19 9 19 19 16 30 10 14 8, 3 3, 1 4, 2 2, 1 3, 1 8, 3 4, 3 4, 3 5, 3 8, 2 4, 2 4, 3 4, 3 8, 3 1, 1 5, 3 2, 1 1, 1 1, 1 2, 1 8, 3 2, 1 1, 1 1, 1 1949 3399 2419 1101 2188 1543 1828 1853 1821 744 1789 1817 1811 1413 1417 3663 760 3875 1717 1043 2223 1059 1289 1140 26.0 45.1 32.1 12.6 28.9 20.6 24.2 24.5 24.1 8.7 23.6 24.0 23.9 18.8 16.8 48.2 8.9 49.6 20.8 11.7 29.7 12.0 15.1 13.2 CG 130.......... CG 142.......... CG 144.......... Mrk 756 ........ CG 150.......... CG 152.......... CG 159.......... UM 483......... CG 165.......... Haro 6........... Mrk 1315 ...... CG 166.......... Mrk 49 .......... Mrk 1321 ...... UM 491......... Wa 52............ CG 173.......... Wa 53............ Mrk 1323 ...... Mrk 51 .......... UM 500......... UM 501......... UM 504......... Mrk 772 ........ CG 184.......... 11 54 36.7 12 00 16.1 12 00 37.0 12 01 27.5 12 04 37.3 12 05 37.8 12 09 36.7 12 12 14.7 12 14 34.3 12 15 19.3 12 15 18.5 12 17 07.3 12 19 09.7 12 19 27.7 12 19 53.1 12 19 56.7 12 21 00.3 12 21 29.5 12 23 54.1 12 24 14.3 12 26 13.2 12 26 23.4 12 32 23.6 12 32 33.4 12 33 06.3 30 42 36 31 13 23 28 59 47 14 02 06 30 51 24 31 56 18 28 45 18 00 04 19 29 55 01 05 45 43 20 38 25 30 38 44 03 51 28 05 02 51 01 46 23 28 33 27 30 37 45 28 22 15 03 05 06 04 13 23
01 18 18
01 15 08
01 44 24 09 10 25 32 05 27 16.64 14.80 16.76 15.20 14.80 17.44 18.20 15.93 19.07 15.20 14.30 15.70 14.50 15.20 15.81 16.10 19.76 16.20 15.40 15.20 14.76 16.53 16.46 15.30 14.10 0.00 0.00 0.00 0.02 0.00 0.00 0.02 0.02 0.03 0.00 0.07 0.02 0.02 0.00 0.01 0.09 0.04 0.10 0.01 0.00 0.04 0.04 0.04 0.00 0.00 24 48 21 40 56 6 5 17 13 24 207 40 22 64 22 24 4 32 24 56 46 35 19 24 44 18 24 10 32 32 6 4 14 4 16 207 24 18 16 17 16 4 32 16 8 46 16 19 24 26 3, 1 4, 3 3, 1 8, 3 4, 3 3, 1 3, 1 1, 1 3, 1 4, 3 8, 2 4, 3 8, 2 8, 3 1, 1 5, 3 3, 1 7, 3 8, 3 8, 3 1, 1 1, 1 1, 1 8, 3 4, 2 3252 615 2960 1479 613 6517 6577 2314 8503 2024 854 778 1525 2009 1994 2571 6416 2578 1874 1189 2033 1992 2009 1154 937 43.2 8.1 39.2 18.5 8.0 86.8 87.5 29.0 113.3 25.4 10.7 10.3 18.7 25.2 24.9 34.1 85.5 34.2 23.4 14.3 25.3 24.7 25.0 14.2 12.6 Mrk 1329 ...... CG 187.......... CG 189.......... UM 513......... Haro 33......... Mrk 1335 ...... Mrk 1338 ...... UM 523......... 12 37 02.3 12 38 30.6 12 40 28.8 12 42 00.6 12 44 38.8 12 46 55.4 12 53 10.1 12 54 51.4 06 55 29 31 53 19 29 23 20 00 45 13 28 28 10 26 33 51 25 16 41 02 39 14 14.60 16.88 18.95 17.08 14.90 15.00 15.50 14.52 0.00 0.02 0.08 0.00 0.09 0.04 0.01 0.05 44 26 10 15 56 40 32 83 31 8 5 12 24 32 24 31 8, 2 3, 1 3, 1 1, 1 4, 3 8, 3 8, 3 1, 1 1635 4302 9219 3755 942 841 1084 932 20.5 57.5 123.0 48.5 12.6 11.1 14.4 11.1 UM 533......... UM 538......... 12 59 57.5 13 02 40.7 02 03 01 01 04 25 14.83 17.70 0.00 0.00 52 17 33 12 1, 1 1, 1 882 904 10.4 10.7 Name (1) R.A. (J2000) (2) Wa 22............ 11 29 14.2 Mrk 424 ........ CG 112.......... Wa 23............ UM 439......... CG 117.......... Mrk 426 ........ Wa 25............ Wa 27............ Wa 28............ Mrk 747 ........ Wa 29............ CG 123.......... CG 124.......... Mrk 429 ........ UM 452......... Wa 34............ Mrk 750 ........ UM 455......... UM 456......... UM 461......... Mrk 641 ........ UM 462......... UM 463......... UM 465......... Decl. (J2000) (3) 195 Other Names (11) IC 700, UGC 6487, Re 1205 Mrk 737 UGC 6561, Ho 267 UGC 6578 Mrk 746 UGC 6655, Ho 275 UGC 6684 Wa 30 Mrk 1307, UGC 6850 Mrk 1308, IC 745, UGC 6877 Mrk 757 Mrk 1313 NGC 4204, UGC 4761 Haro 8, UGC 7354 UGC 7531B UGC 7704, Mrk 773, Re 2510 IC 3589, UGC 7790 NGC 4810, A277, UGC 8034 UGC 8105 196 SALZER ET AL. TABLE 2—Continued Name (1) R.A. (J2000) (2) Decl. (J2000) (3) B (4) AB (5) a25 (arcsec) (6) b25 (arcsec) (7) Refs. (8) V (km s
1) (9) Dist. (Mpc) (10) UM 539......... Mrk 450 ........ Mrk 786 ........ UM 559......... UM 562......... Wa 69............ Haro 38......... Mrk 67 .......... UM 618......... Wa 81............ Mrk 1369 ...... Mrk 675 ........ Mrk 1384 ...... Haro 43......... Mrk 475 ........ Mrk 829 ........ 13 02 52.0 13 14 48.3 13 16 52.2 13 17 43.8 13 18 19.0 13 25 48.9 13 35 35.6 13 41 56.2 13 52 36.7 13 57 09.5 14 04 14.8 14 19 14.4 14 32 52.8 14 36 07.9 14 39 04.8 14 50 56.4 00 24 24 34 52 43 12 32 55
01 00 00 02 12 42 33 03 49 29 12 58 30 31 10 00 01 22 29 13 17 36 43 32 36 21 39 05 59 56 28 26 54 36 48 13 35 34 17 18.88 14.90 14.60 15.81 19.93 16.90 15.30 16.31 17.96 16.90 17.00 16.50 17.00 15.40 16.31 14.50 0.00 0.00 0.01 0.02 0.00 0.01 0.02 0.00 0.06 0.00 0.00 0.00 0.05 0.00 0.00 0.00 7 48 44 44 7 16 48 19 11 16 40 16 16 40 56 35 6 35 31 20 6 8 11 16 10 8 16 16 16 16 21 11 1, 1 8, 2 8, 2 1, 1 1, 1 5, 3 4, 2 2, 1 1, 1 5, 3 8, 3 8, 3 8, 3 4, 3 2, 1 8, 2 6236 860 967 1233 5426 4363 853 968 4499 2260 3571 3058 2274 1908 540 1198 81.8 12.1 12.3 15.1 71.2 58.8 11.9 13.6 59.1 30.9 48.8 42.1 30.2 26.5 8.7 17.5 II Zw 71 ........ Mrk 1390 ...... Mrk 850 ........ Mrk 689 ........ Mrk 900 ........ 14 51 14.3 15 01 02.9 15 22 14.6 15 36 19.2 21 29 59.4 35 32 28 00 42 30 31 28 31 30 40 55 02 24 50 14.40 15.60 16.50 15.20 14.59 0.00 0.14 0.01 0.07 0.17 48 32 16 32 44 48 16 16 24 36 4, 2 8, 3 8, 3 8, 3 2, 1 1253 1679 2197 1730 1148 18.3 22.3 31.0 24.8 18.1 Mrk 324 ........ Mrk 328 ........ 23 26 32.1 23 37 39.0 18 15 57 30 07 46 15.38 15.46 0.08 0.24 29 28 23 25 2, 1 2, 1 1599 1385 24.4 21.8 Other Names (11) UGC 8323, VV 616 NGC 5058, UGC 8345 UGC 8578 UGC 9560, II Zw 70, VV 324b UGC 9562, VV 324 NGC 7077, UGC 11755 Note.—Units of right ascension are hours, minutes, and seconds, and units of declination are degrees, arcminutes, and arcseconds. 7. Minor-axis diameter measured at the B = 25.0 mag arcsec
2 level, in arcseconds. 8. References for the apparent magnitude (left number) and diameter measurements (right number) (see x 2.5). 9. Observed heliocentric radial velocity, obtained from the 21 cm observations (kilometers per second), where we adopt the normal method of using the ‘‘ optical ’’ convention for measuring the Doppler shift. If the galaxy was not detected in H i, the velocity listed is from published optical spectroscopy. In all cases, the optical velocity reference is the same as the apparent magnitude reference in column (8). Velocity uncertainties for the H i–detected galaxies depend on the velocity resolution used for the observations. A conservative estimate of the velocity uncertainty would be 4 km s
1 for galaxies observed with the highest velocity resolution (R = 1 in Table 3), 8 km s
1 for galaxies observed with intermediate resolution (R = 2 in Table 3), and 16 km s
1 for galaxies observed with the lowest resolution (R = 3 in Table 3). These estimated uncertainties, equal to two channel widths, are consistent with the observed velocity differences between our sample and that of Smoker et al. (2000) for 19 galaxies in common. 10. Distance to each galaxy, in megaparsecs (see x 2.6). 11. Alternate designations for galaxies. These have been limited to NGC/IC and UGC designations, plus identifications from the other objective-prism surveys when the object is included in more than one of the surveys. Of the 139 objects observed, a total of 122 were detected at Arecibo. Table 3 lists the data obtained from the H i observations, plus a number of quantities derived from the data. Details of the computations of these various derived quantities are given in x 2.6. The contents of Table 3 include the following: 1. Galaxy name, repeated from Table 2. 2. Blue absolute magnitude (H0 = 75 km s
1 Mpc
1 is assumed), corrected for Galactic absorption (AB) but not for internal absorption. 3. Width of the H i profile, measured at 50% of the peak flux (km s
1). 4. Width of the H i profile, measured at 20% of the peak flux (km s
1).
1 Ð 5. The H i flux integral F.I. (Jy km s ), where F.I. = S dv and S is the 21 cm flux density in janskys. 6. The rms scatter in the baseline fit of the 21 cm scan (mJy). 7. An instrumental code specifying the 21 cm observational setup used. The first number (‘‘ F ’’) indicates the feed used [(1) 21 cm feed; (4) 22 cm feed], and the second number (‘‘ R ’’) indicates the channel spacing (and hence velocity resolution) of the data [(1) 2 km s
1 per channel; (2) 4 km s
1 per channel; (3) 8 km s
1 per channel]. 8. The number of on-off pairs obtained. 9. Logarithm of the H i mass (M). 10. H i mass–to–blue light ratio [(M/L)]. 11. Logarithm of the total (dynamical) mass (M). 12. Total mass–to–blue light ratio [(M/L)]. For the 17 galaxies not detected in H i, the obvious entries are left blank. The values listed for the H i flux integral, H i mass, and H i mass–to–blue light ratio for the nondetections are 3  upper limits, computed assuming a profile width at zero intensity of 120 km s
1. The H i profiles for all detected galaxies are shown in Figure 1. When examining the plots, it is important to remember that three different velocity ranges are used, corresponding to the three bandwidths employed. Note the variety in the shapes of the profiles. The majority of TABLE 3 H i Observations Name (1) MB (2) W50 (3) W20 (4) F.I. (5) rms (6) F, R (7) No. (8) log MH i (9) MH i/LB (10) log Mtot (11) Mtot/LB (12) UM 241......... UM 38 .......... UM 40 .......... UM 51 .......... UM 69 .......... UM 80 .......... UM 92 .......... Mrk 1153 ...... UM 102......... UM 323......... UM 330......... UM 112......... UM 336......... UM 345......... UM 133......... UM 372......... UM 382......... UM 396......... UM 408......... UM 417......... Mrk 1038 ...... Mrk 370 ........ Mrk 600 ........ CG 4 ............. II Zw 40 ........ CG 6 ............. CG 10 ........... CG 13 ........... CG 14 ........... CG 23 ........... CG 30 ........... CG 31 ........... Mrk 408 ........ CG 34 ........... Haro 22......... Wa 4 ............. Mrk 411 ........ CG 46 ........... Mrk 714 ........ CG 50 ........... CG 55 ........... Wa 5 ............. Wa 6 ............. Wa 8 ............. Wa 13............ Mrk 724 ........ Mrk 416 ........ Mrk 1263 ...... Mrk 1264 ...... Mrk 1271 ...... CG 75 ........... CG 76 ........... CG 80 ........... CG 82 ........... Mrk 1283 ...... UM 422......... CG 101.......... CG 103.......... Wa 22............ Mrk 424 ........ CG 112.......... Wa 23............ UM 439......... CG 117..........
16.88
16.08
16.06
16.85
16.89
17.15
16.40
16.33
15.98
16.17
16.17
15.30
16.35
16.56
15.99
16.75
15.00
16.34
15.86
14.58
16.93
17.13
15.41
16.55
16.17
16.88
14.45
16.10
16.36
16.23
17.36
16.96
16.73
16.07
16.05
16.70
16.61
16.28
15.65
16.67
14.76
13.58
16.76
15.37
15.78
14.37
15.98
15.66
15.00
15.64
14.60
14.65
15.94
13.95
14.44
17.40
16.49
16.43
16.25
16.88
16.72
17.03
15.40
15.99 95 107 97 151 141 40 96 ... 116 52 61 75 151 112 72 66 ... ... 65 46 ... 112 73 ... 99 117 88 138 108 105 203 49 80 60 69 153 168 ... 50 61 112 48 195 46 148 42 85 125 146 96 75 75 182 41 ... 101 156 104 90 96 135 105 87 ... 137 140 114 185 160 67 115 ... 152 85 96 109 167 126 110 101 ... ... 88 64 ... 150 96 ... 144 150 134 156 125 135 238 71 134 88 95 213 193 ... 92 107 125 69 217 83 169 75 164 143 172 141 100 125 199 66 ... 134 196 138 137 134 160 163 118 ... 0.673 2.285 5.903 1.495 4.632 1.448 0.177 <0.212 2.161 1.404 0.402 3.500 0.518 1.265 3.891 2.915 <0.275 <0.170 1.194 0.424 <0.202 8.431 5.901 <0.249 10.010 1.231 0.691 5.896 0.795 0.901 0.593 0.546 2.329 0.200 0.858 6.802 3.421 <0.107 0.930 1.067 1.464 0.279 1.360 2.148 32.410 0.164 1.233 3.781 3.281 0.503 0.828 0.369 33.890 1.147 <0.117 15.110 3.821 2.986 5.797 1.702 1.490 2.000 5.372 <0.249 0.82 1.30 1.04 1.08 1.01 0.84 0.74 1.32 1.10 1.83 0.82 1.17 0.90 0.98 1.28 1.52 1.71 1.06 0.85 1.20 1.26 1.39 1.24 1.55 1.29 0.89 0.81 1.08 1.23 0.87 0.56 1.05 1.28 0.67 0.78 1.15 1.45 0.67 0.79 0.93 0.75 0.67 0.78 1.49 1.52 0.71 0.68 1.30 1.08 0.64 0.59 1.01 0.75 0.93 0.73 1.59 0.89 0.76 1.06 1.45 0.76 1.13 2.02 1.55 4, 3 1, 1 1, 1 4, 2 1, 2 4, 1 4, 3 1, 3 1, 1 1, 1 4, 3 1, 1 4, 3 4, 2 1, 1 1, 1 1, 3 4, 3 4, 1 1, 2 1, 3 4, 1 4, 1 4, 3 1, 1 4, 2 4, 2 1, 2 1, 3 1, 1 4, 3 1, 2 1, 1 4, 3 1, 1 1, 2 1, 1 4, 3 1, 1 1, 1 1, 1 1, 3 4, 3 1, 1 1, 1 1, 3 1, 1 1, 1 1, 2 1, 2 1, 2 1, 3 1, 2 1, 1 1, 3 1, 1 4, 2 1, 2 1, 1 1, 2 4, 2 1, 2 1, 1 4, 3 3 4 5 3 2 5 4 2 6 4 3 3 3 4 3 3 4 2 8 6 2 3 6 1 4 4 6 2 4 6 5 3 4 3 6 2 4 2 4 4 7 3 2 2 2 2 5 2 2 4 4 2 2 3 2 2 4 2 2 4 4 2 4 2 8.759 8.423 8.829 9.199 9.068 9.192 8.573 <7.626 8.685 8.414 8.669 9.030 8.877 9.255 8.845 8.771 <8.234 <8.455 8.815 8.102 <7.589 8.718 8.523 <7.648 8.362 8.335 8.028 9.168 8.125 8.216 9.164 7.843 8.428 8.351 7.940 8.669 8.541 <8.242 7.745 7.962 8.123 7.246 9.214 8.027 8.840 6.958 8.085 8.504 7.721 7.300 7.749 7.610 9.350 7.331 <6.532 9.640 8.894 8.576 8.827 8.560 8.869 8.788 8.400 <7.720 0.68 0.66 1.70 1.94 1.38 1.45 0.69 <0.08 1.33 0.59 1.07 5.45 1.47 2.86 1.88 0.79 <1.15 <0.56 2.00 1.25 <0.04 0.50 1.53 <0.07 0.53 0.26 1.19 3.60 0.26 0.36 1.12 0.08 0.37 0.56 0.22 0.66 0.53 <0.36 0.21 0.13 1.12 0.44 2.17 0.51 2.27 0.11 0.33 1.17 0.35 0.07 0.55 0.38 6.33 0.38 <0.04 3.22 1.33 0.68 1.43 0.43 1.02 0.64 1.17 <0.14 10.043 9.517 9.408 10.364 9.981 9.342 9.578 ... 9.635 8.837 9.136 9.617 10.271 9.657 9.370 9.350 ... ... 9.074 8.571 ... 9.628 9.013 ... 9.039 9.595 9.083 10.099 9.605 9.800 10.257 8.859 9.262 8.961 9.116 9.766 9.880 ... 8.712 8.964 9.340 8.451 10.452 8.556 9.655 8.741 9.412 9.659 9.324 9.190 9.313 8.848 10.221 8.345 ... 9.966 9.727 9.510 9.486 9.596 9.749 9.721 9.146 ... 13.19 8.14 6.47 28.31 11.33 2.04 6.98 ... 11.83 1.58 3.15 21.08 36.28 7.22 6.31 3.00 ... ... 3.62 3.69 ... 4.03 4.74 ... 2.51 4.67 13.51 30.81 7.78 13.67 13.85 0.80 2.51 2.29 3.33 8.21 11.57 ... 1.91 1.33 18.40 7.02 37.60 1.72 14.83 6.62 7.03 16.65 14.19 5.77 19.99 6.57 47.06 3.93 ... 6.84 9.07 5.86 6.51 4.72 7.75 5.44 6.53 ... TABLE 3—Continued Name (1) MB (2) W50 (3) W20 (4) F.I. (5) rms (6) F, R (7) No. (8) log MH i (9) MH i/LB (10) log Mtot (11) Mtot/LB (12) Mrk 426 ........ Wa 25............ Wa 27............ Wa 28............ Mrk 747 ........ Wa 29............ CG 123.......... CG 124.......... Mrk 429 ........ UM 452......... Wa 34............ Mrk 750 ........ UM 455......... UM 456......... Mrk 641 ........ UM 463......... UM 465......... CG 130.......... CG 142.......... CG 144.......... Mrk 756 ........ CG 150.......... CG 152.......... CG 159.......... UM 483......... CG 165.......... Haro 6........... Mrk 1315 ...... CG 166.......... Mrk 49 .......... Mrk 1321 ...... UM 491......... Wa 52............ CG 173.......... Wa 53............ Mrk 1323 ...... Mrk 51 .......... UM 500......... UM 501......... UM 504......... Mrk 772 ........ CG 184.......... Mrk 1329 ...... CG 187.......... CG 189.......... UM 513......... Haro 33......... Mrk 1335 ...... Mrk 1338 ...... UM 523......... UM 533......... UM 538......... UM 539......... Mrk 450 ........ Mrk 786 ........ UM 559......... UM 562......... Wa 69............ Haro 38......... Mrk 67 .......... UM 618......... Wa 81............ Mrk 1369 ...... Mrk 675 ........ Mrk 1384 ......
16.68
16.42
16.45
15.21
15.24
16.40
16.43
16.93
16.38
15.62
16.57
14.35
16.42
16.12
15.86
13.07
16.49
16.53
14.73
16.21
16.16
14.73
17.25
16.53
16.40
16.23
16.83
15.92
14.39
16.88
16.81
16.18
16.65
14.94
16.57
16.45
15.58
17.29
15.48
15.57
15.46
16.41
16.96
16.94
16.58
16.35
15.69
15.28
15.30
15.75
15.26
12.45
15.68
15.52
15.86
15.11
14.33
16.96
15.10
14.35
15.96
15.55
16.44
16.62
15.45 117 69 104 98 48 138 75 100 182 68 59 50 146 79 42 ... 79 76 104 181 137 39 211 78 92 113 82 76 66 46 ... 63 123 ... 78 67 173 92 63 ... 64 62 83 180 69 59 50 56 33 145 48 42 ... 64 159 104 ... 94 83 62 51 52 169 ... 154 179 113 171 140 87 158 117 142 217 75 68 70 168 113 88 ... 141 97 118 281 158 81 262 129 142 174 130 94 75 67 ... 84 151 ... 123 88 194 125 73 ... 96 92 123 202 125 84 88 102 60 176 64 66 ... 87 189 125 ... 184 128 83 76 77 275 ... 175 4.435 2.101 3.615 5.007 1.063 5.359 1.224 2.542 1.300 0.379 0.131 1.242 1.692 3.243 0.451 <0.346 1.242 0.643 2.050 0.984 4.349 1.177 1.487 0.241 1.593 0.269 2.072 13.070 0.109 2.218 <0.212 0.859 3.436 <0.175 0.423 1.011 0.997 5.541 2.018 <0.400 1.670 4.002 6.864 1.649 0.536 0.482 5.551 0.678 0.792 13.790 2.269 0.172 <0.286 3.419 10.690 4.190 <0.202 0.709 2.297 0.545 0.212 0.458 1.058 <0.206 3.872 1.53 0.98 0.94 1.52 0.95 0.85 0.77 0.84 1.71 2.72 0.88 0.96 2.93 2.44 1.44 2.15 1.36 0.56 0.94 0.81 0.67 0.78 0.67 0.67 1.14 0.52 1.21 1.03 0.52 1.35 1.32 1.31 1.17 1.09 0.75 1.08 1.35 2.75 2.65 2.49 0.88 1.66 1.19 0.89 0.90 1.10 1.43 0.74 0.68 1.89 2.04 1.00 1.78 2.25 1.36 1.71 1.26 0.93 0.83 0.60 1.08 1.00 1.28 1.28 1.26 1, 2 1, 1 1, 2 1, 1 1, 1 1, 2 1, 2 1, 2 1, 3 1, 2 4, 3 1, 1 1, 3 1, 2 4, 2 1, 3 1, 2 4, 2 1, 1 4, 3 1, 2 1, 1 4, 3 4, 3 4, 2 4, 3 4, 1 1, 1 1, 3 1, 1 4, 3 4, 1 4, 2 4, 3 4, 3 1, 2 1, 3 1, 2 1, 1 1, 3 1, 1 1, 1 1, 1 4, 3 4, 3 4, 3 1, 1 1, 1 1, 1 1, 1 1, 1 1, 3 4, 3 1, 1 1, 1 1, 2 4, 3 4, 3 1, 1 1, 1 4, 3 4, 1 1, 3 1, 3 4, 2 2 3 2 2 4 2 2 2 1 1 2 3 2 2 4 1 5 5 4 3 2 4 3 2 6 3 4 2 3 2 1 6 2 1 3 3 1 2 3 1 4 2 2 2 2 2 2 4 6 2 2 3 1 2 2 4 1 2 5 8 2 8 3 3 2 8.772 8.524 8.787 8.921 7.313 8.979 8.283 8.636 8.114 7.499 7.882 7.423 9.011 8.627 8.014 <7.287 7.744 8.496 7.623 8.580 8.668 7.427 9.426 8.641 8.525 8.920 8.543 9.609 6.538 8.305 <7.668 8.138 9.017 <8.482 8.165 8.157 7.810 9.104 8.535 <7.808 7.957 8.297 8.964 9.147 9.288 8.447 8.470 7.419 7.664 8.862 7.927 6.689 <8.659 8.227 8.712 8.459 <8.388 8.779 7.988 7.405 8.254 8.029 8.859 <7.961 8.947 0.85 0.61 1.09 4.63 0.11 1.77 0.35 0.49 0.25 0.12 0.12 0.32 1.87 1.01 0.31 <0.77 0.09 0.51 0.36 0.84 1.08 0.23 2.25 0.72 0.62 1.80 0.44 11.75 0.04 0.24 <0.06 0.31 1.52 <2.16 0.23 0.25 0.26 1.03 1.49 <0.26 0.40 0.36 1.02 1.58 3.05 0.54 1.05 0.14 0.24 2.45 0.45 0.35 <1.64 0.71 1.57 1.75 <3.04 0.67 0.60 0.31 0.50 0.43 1.29 <0.14 3.94 9.666 9.101 9.575 9.476 8.211 9.916 9.170 9.551 9.946 9.044 9.547 8.399 9.799 9.281 8.699 ... 8.851 9.423 9.157 9.959 9.825 8.394 10.484 9.106 9.320 9.769 9.213 10.226 8.825 8.636 ... 8.998 9.692 ... 9.942 9.001 9.876 10.112 9.059 ... 9.263 8.896 9.413 10.180 9.302 9.084 8.762 8.826 8.345 9.791 8.681 8.126 ... 9.010 9.755 9.380 ... 9.455 9.084 8.712 9.021 8.660 10.256 ... 10.177 6.66 2.30 6.66 16.61 0.88 15.31 2.67 4.03 16.75 4.21 5.56 3.06 11.48 4.58 1.52 ... 1.21 4.33 12.36 20.16 15.43 2.15 25.72 2.10 3.87 12.73 2.04 48.55 7.90 0.51 ... 2.26 7.22 ... 13.85 1.77 29.68 10.51 4.97 ... 8.07 1.45 2.86 17.07 3.15 2.36 2.06 3.50 1.13 20.79 2.54 9.45 ... 4.28 17.35 14.59 ... 3.16 7.48 6.31 2.92 1.86 32.15 ... 66.90 H i IN STAR-FORMING DWARF GALAXIES 199 TABLE 3—Continued Name (1) MB (2) W50 (3) W20 (4) F.I. (5) rms (6) F, R (7) No. (8) log MH i (9) MH i/LB (10) log Mtot (11) Mtot/LB (12) Haro 43......... Mrk 475 ........ Mrk 829 ........ II Zw 71 ........ Mrk 1390 ...... Mrk 850 ........ Mrk 689 ........ Mrk 900 ........ Mrk 324 ........ Mrk 328 ........
16.72
13.39
16.72
16.91
16.28
15.96
16.85
16.87
16.63
16.47 125 ... 77 173 ... 56 150 96 106 115 150 ... 133 191 ... 93 179 142 180 151 4.098 <0.178 2.709 6.276 <0.193 0.591 4.621 1.929 1.802 0.801 1.23 1.11 1.30 1.45 1.20 0.84 0.89 1.27 0.93 0.91 1, 2 1, 3 1, 1 1, 1 1, 3 4, 2 1, 2 4, 1 4, 1 4, 1 2 3 7 5 3 5 3 6 7 1 8.917 <6.648 8.356 8.888 <7.418 8.153 8.905 8.317 8.469 8.023 1.14 <0.13 0.31 0.90 <0.05 0.39 0.99 0.25 0.44 0.18 9.729 ... 9.056 10.537 ... 9.310 9.898 9.561 9.575 9.678 7.41 ... 1.57 40.01 ... 5.65 9.73 4.39 5.62 8.26 the galaxies have a single-peaked profile, although a few double-horned profiles (indicative of a rotating disk system) are present as well. The single-peaked profiles range from roughly rectangular (e.g., Was 13, Mrk 1263), to Gaussian (e.g., UM 80, CG 82 = Mrk 36), to pyramidal (e.g., Haro 38, Mrk 900). One of the apparently double-horned cases, UM 523, is in fact an interacting pair of dwarf galaxies that were both in the Arecibo beam; the observed profile is the superposition of the two roughly Gaussian H i profiles. While most of the H i profiles shown in Figure 1 are quite strong and possess high signal-to-noise ratios, there are a number of examples of weak detections as well (e.g., UM 92, UM 330, and UM 417 in the first page of Fig. 1). In at least some cases, the reality of these marginal detections might well be questioned. However, we remind the reader that all the galaxies in our sample have optical redshifts, which are usually of high quality because these objects typically exhibit strong emission-line spectra. Therefore, the close agreement between the optical velocity with the putative H i signal was taken as evidence for the reality of the latter. 2.4. Comparison with Previous Observations As mentioned above, many studies of the H i characteristics of dwarf star-forming galaxies have been carried out in the past. Here we compare our data for galaxies in common with those of the previous studies. The largest overlap is with the study of Thuan & Martin (1981), for which we have 27 objects in common (24 of which were detected in both surveys), and Smoker et al. (2000), for which there are 29 galaxies in common (19 of which were detected in both surveys). In general, we find good agreement with the Smoker et al. data set for the observed H i parameters (velocity, line widths, and fluxes). However, we note that they failed to detect a number of galaxies for which we were able to obtain reliable signals (UM 417, 452, 455, 465, 491, 513, and 538). The agreement between the Thuan & Martin values and ours was not as good. The biggest difference between the two data sets is the noise levels obtained. While the current Arecibo data have a typical rms noise level of 1 mJy, the noise in the Thuan & Martin data is typically 5–15 mJy. The higher noise levels in the older data necessarily lead to higher uncertainties in the measured quantities. Both velocity widths and total H i fluxes are affected. We have 11 galaxies in common with Gordon & Gottesman (1981), two in common with Staveley-Smith et al. (1992), and three in common with Thuan et al. (1999). In all but one case, our data are in good agreement with those of these previous studies. The one exception is Mrk 475, for which Thuan et al. claim a weak detection, while we do not detect this source at all. Our spectrum has an rms noise level of 1.11 mJy, and we should have easily detected a signal of the strength seen by Thuan et al. (1999). Their spectral profile is quite narrow (less than three channels FWHM), suggesting the possibility that it is radio-frequency interference rather than a real detection. Further observations of this source will be required to clarify the situation. 2.5. Optical Data In addition to the newly acquired H i data, we include in Table 2 optical data for the sample galaxies. Much of this comes from previously published work by the authors. Apparent magnitudes.—The magnitude information comes from eight different sources, and the origin of the data is specified in column (8) (left number) of the table. The reference numbers refer to the following: (1) CCD photometry from Salzer et al. (1989a) for the UM ELGs (N = 27); (2) CCD photometry of Norton & Salzer (2002) for extreme BCDs (N = 14); (3) CCD photometry from Salzer et al. (1995) for the Case galaxies (N = 21); (4) magnitudes from the Catalogue of Galaxies and of Clusters of Galaxies (CGCG; Zwicky et al. 1961–1968), corrected as described in Salzer (1989; N = 27); (5) CCD photometry from Bothun et al. (1989) for the Wasilewski sample, converted from R band to B band assuming B
R = 0.4 (N = 7); (6) estimated magnitudes from the original UM survey lists (MacAlpine & Williams 1981), corrected as described in Salzer et al. (1989a; N = 6); (7) estimated magnitudes from the original Wasilewski (1983) survey list (N = 1); and (8) magnitudes taken from the compilation of Mazzarella & Balzano (1986) for the Markarian galaxies (N = 36). Fully 50% of the sample possess modern photometry based on CCD images. Angular diameters.—The angular diameters listed in Table 2 come from one of three sources, as specified in column (8) (right number). These sources are (1) measurements from CCD images, taken at the B = 25.0 mag arcsec
2 level (N = 63); (2) blue diameters listed in the UGC (N=24); and (3) diameters measured directly from the Palomar Observatory Sky Survey (POSS) prints by us (N = 54). For both sources 2 and 3, the diameter measurements were put on the same isophotal system as the CCD diameters by comparing sizes measured for galaxies that also had CCD images. That is, we compared sizes for the subset of galaxies with both UGC diameters and available CCD images and also measured the sizes for the same galaxies on the POSS. In the case of the UGC diameters, we found that the cataloged sizes 24 UM 241 6 16 60 UM 38 UM 40 40 3 8 20 0 0 0 -3 -8 2400 3200 4000 4800 5600 6400 1000 1200 1400 1600 1800 36 UM 51 12 30 UM 69 24 12 0 4400 4800 5200 1400 1600 1800 UM 80 0 -6 4000 1200 10 0 3600 1000 20 6 3200 -20 800 1200 1600 2000 2400 2800 24 4400 4600 4800 5000 5200 30 4 UM 92 16 UM 102 20 UM 323 2 8 10 0 0 Flux [mJy] 0 -2 -8 4800 5600 6400 7200 8000 8800 1800 2000 2200 2400 2600 -10 1400 1600 1800 2000 2200 2400 6400 7200 8000 1800 2000 2200 45 UM 330 6 30 4 UM 112 UM 336 2 3 15 0 0 0 -2 -3 -15 3200 4000 4800 5600 6400 1600 1800 2000 2200 2400 4000 4800 5600 45 12 UM 345 45 UM 133 UM 372 30 30 6 15 15 0 0 0 -6 4800 24 16 5200 5600 6000 6400 6800 1200 1400 1600 1800 2000 12 UM 408 8 1200 1400 1600 70 UM 417 MK 370 50 4 8 30 0 0 0 -4 -8 3200 3400 3600 3800 4000 2000 2400 2800 3200 3600 400 600 800 1000 1200 Velocity [km/s] Fig. 1.—Observed H i profiles for 122 detected dwarf star-forming galaxies. The vertical axis has units of millijanskys, while the horizontal axis is recession velocity in kilometers per second. Note that the velocity coverage varies depending on the bandwidth used; the displayed velocity range is either 1000, 2000, or 4000 km s
1. A wide range of profile shapes is observed, from standard double-horned to narrow Gaussian and even pyramidal shapes. This indicates that a variety of dynamical processes and gas distributions are likely to be present in this sample. 200 70 90 MK 600 12 II ZW 40 50 60 8 30 30 4 0 0 0 600 800 1000 1200 1400 400 9 600 800 1000 CG 6 1200 1200 1600 2000 2400 2800 45 CG 10 CG 13 6 CG 14 6 30 3 3 15 0 0 0 -3 -3 1200 1600 2000 2400 2800 1200 1600 2000 2400 2800 12 8 0 800 1600 2400 3200 4000 12 4 CG 23 CG 30 CG 31 8 2 Flux [mJy] 4 4 0 0 0 -2 -4 1200 -4 1400 1600 1800 2000 -4 2200 30 6400 7200 8000 8800 9600 800 6 MK 408 CG 34 20 1200 1600 2000 2400 1400 1600 1800 15 HARO 22 10 3 5 10 0 0 0 -3 -5 1000 1200 1400 1600 1800 2000 3200 4000 4800 5600 6400 1000 1200 24 45 WAS 4 16 MK 411 MK 714 15 30 10 8 15 5 0 0 0 400 800 1200 1600 -8 1000 2000 16 15 CG 50 1200 1400 1600 1800 2000 CG 55 1000 1200 1400 1600 1800 1600 2400 3200 4000 WAS 5 6 10 8 800 9 5 3 0 0 0 -8 -5 800 1000 1200 1400 1600 1800 -3 1000 1200 1400 1600 Velocity [km/s] Fig. 1.—Continued 201 1800 0 800 8 WAS 6 36 WAS 8 120 24 4 60 12 0 WAS 13 180 0 0 -4 -60 4000 4800 5600 6400 7200 600 800 1000 1200 1400 0 200 400 600 800 1000 1400 1600 1800 6 MK 724 12 3 MK 416 MK 1263 24 8 12 4 0 0 0 -3 -800 24 0 800 1600 2400 3200 MK 1264 -4 800 8 6 1000 1200 1400 1600 1800 1000 12 MK 1271 1200 CG 75 8 4 16 -12 800 2 4 8 Flux [mJy] 0 0 0 -2 0 400 800 1200 1600 400 800 1200 1600 2000 400 800 1200 1600 2000 600 800 1000 9 CG 76 180 CG 80 CG 82 24 6 16 120 3 8 60 0 0 0 -3 0 800 1600 2400 3200 150 0 400 800 1200 1600 30 UM 422 200 CG 101 CG 103 100 20 20 50 10 10 0 0 0 1200 1400 1600 1800 2000 1600 2000 2400 2800 3200 24 60 WAS 22 400 30 800 1200 1600 2000 2400 16 MK 424 12 16 40 CG 112 8 8 20 4 0 0 0 1000 1200 1400 1600 1800 1200 1600 2000 2400 Velocity [km/s] Fig. 1.—Continued 202 2800 2400 2800 3200 3600 4000 4400 36 WAS 23 16 UM 439 60 MK 426 24 40 8 12 20 0 0 1600 2000 2400 2800 3200 0 600 800 1000 1200 1400 1600 800 1200 1600 2000 2400 60 30 WAS 25 20 WAS 27 WAS 28 40 20 10 0 1400 1600 1800 2000 2200 10 20 0 0 800 1200 1600 2000 2400 2800 1400 1600 1800 2000 2200 24 MK 747 WAS 29 45 CG 123 15 16 10 30 8 5 Flux [mJy] 15 0 0 0 400 600 800 1000 1200 10 CG 124 24 800 1200 1600 2000 2400 2800 MK 429 800 12 1200 1600 2000 2400 2800 UM 452 8 5 16 4 0 8 0 -5 0 -4 800 12 1200 1600 2000 2400 WAS 34 4 0 2400 0 3200 4000 800 1600 2400 3200 0 40 MK 750 24 8 1600 2800 16 20 8 0 0 -20 4800 200 400 600 800 800 1600 2400 3200 UM 455 1000 2400 3200 4000 4800 5600 18 9 UM 456 36 MK 641 UM 465 12 6 24 3 6 12 0 0 0 -3 800 1200 1600 2000 2400 1200 1600 2000 2400 2800 Velocity [km/s] Fig. 1.—Continued 203 3200 0 400 800 1200 1600 2000 9 24 CG 130 6 16 3 8 CG 142 CG 144 6 3 0 0 0 -3 -3 2400 2800 3200 3600 4000 200 MK 756 36 24 24 16 12 8 0 0 400 800 1200 1600 2000 2400 400 600 800 1000 1600 CG 150 4800 0 -3 400 600 800 1000 4800 3 UM 483 2 4000 3 200 CG 159 3200 CG 152 6 18 4 2400 5600 6400 7200 8000 CG 165 2 12 1 0 6 0 Flux [mJy] -2 0 -1 -4 4800 24 5600 6400 7200 8000 1600 180 HARO 6 2000 2400 2800 3200 6400 4 MK 1315 16 120 2 8 60 0 0 0 -2 1600 45 1800 2000 2200 2400 400 15 MK 49 30 600 800 7200 1000 1200 1400 -800 10 20 5 10 0 0 15 8800 9600 10400 CG 166 30 UM 491 8000 0 800 1600 2400 WAS 52 0 1200 30 1400 1600 1800 2000 WAS 53 -5 1400 15 1600 1800 2000 2200 2400 8 MK 1323 10 15 2000 2400 2800 3200 800 1600 2400 MK 51 4 5 0 1600 0 0 -15 -4 800 1600 2400 3200 4000 1200 1600 2000 2400 Velocity [km/s] Fig. 1.—Continued 204 2800 -800 0 30 60 UM 500 UM 501 30 MK 772 20 40 15 10 20 0 0 0 1200 1600 2000 2400 2800 1400 1600 1800 2000 2200 2400 800 1000 1200 1400 1600 12 70 60 CG 184 MK 1329 40 50 20 30 CG 187 8 4 0 0 0 -4 400 600 800 1000 1200 1000 8 1200 1600 1800 2000 2400 10 UM 513 4 4800 5600 60 0 -4 4000 HARO 33 5 0 3200 90 CG 189 Flux [mJy] 1400 30 -5 0 -8 7200 8000 8800 9600 -10 1600 10400 2400 3200 4000 4800 5600 600 800 1000 1200 1400 1000 1200 1400 120 12 18 MK 1335 MK 1338 UM 523 80 12 8 6 4 40 0 0 0 400 600 800 1000 1200 1400 600 800 1000 1200 1400 8 UM 533 40 UM 538 6 800 MK 450 40 4 20 600 60 2 20 0 0 0 -2 400 600 800 1000 1200 -800 0 800 1600 2400 400 600 800 1000 1200 4000 4800 5600 70 45 50 MK 786 UM 559 WAS 69 6 30 3 30 15 0 0 0 -3 -30 600 800 1000 1200 1400 400 800 1200 1600 Velocity [km/s] Fig. 1.—Continued 205 2000 2400 3200 206 SALZER ET AL. Vol. 124 12 30 HARO 38 6 MK 67 9 UM 618 4 20 10 0 400 12 600 800 1000 6 2 3 0 0 -2 -3 600 1200 800 8 WAS 81 8 1000 1200 1400 1600 MK 1369 -4 2400 36 24 3200 4000 4800 5600 MK 1384 4 12 4 0 0 0 -4 -4 1800 2000 2200 2400 2600 2800 2400 3200 4000 4800 HARO 43 MK 829 30 20 24 1600 2000 2400 2800 3200 0 0 0 1200 1600 2000 2400 2800 II ZW 71 15 10 12 800 1000 1200 1400 1600 12 8 -12 1200 30 36 Flux [mJy] 1600 800 1000 1200 1400 1600 1200 1400 1600 30 30 MK 850 MK 689 20 MK 900 20 4 10 10 0 -4 1200 0 0 -10 1600 2000 2400 2800 3200 800 1200 1600 2000 2400 1200 1400 1600 800 1000 9 12 MK 324 MK 328 6 6 3 0 0 -6 1000 1200 1400 1600 1800 2000 800 1000 Velocity [km/s] Fig. 1.—Continued needed to be reduced by a factor of 1.36  0.11 to put them on the same scale as the CCD diameters. The POSS sizes needed to be multiplied by 1.20  0.09 to bring them in line with the CCD data. The uncertainties quoted here are the formal errors in the mean values; the rms scatter about the mean values is 0.43 and 0.41, respectively. The optical data can be used to illustrate the properties of our sample of galaxies. Figure 2 presents a series of histograms showing the distributions of apparent magnitudes, diameters, and absolute magnitudes. Figure 2a reveals a broad magnitude distribution with a median value of mB = 15.7. Fully 50% of the sample are fainter than the CGCG limit. The detailed shape of the magnitude distribution is due to the combination of several surveys with differing depths. Most of the early surveys (Markarian, Haro, Wasilewski) have completeness limits around mB = 15, which accounts for the peak in Figure 2a at this value. The extended tail of objects with mB > 16 is due primarily to the UM and Case surveys. Note that most of the undetected galaxies have mB > 15.5. The diameter distribution shown in Figure 2b is roughly Gaussian, with a peak near 3 kpc and a tail containing a modest number of galaxies extending beyond 6 kpc. The median value is 3.2 kpc. The majority of nondetections rep- No. 1, 2002 30 Full Sample (N = 139) Non-detections (N = 17) 20 15 10 5 0 12 25 14 (b) 16 Median d = 3.2 kpc 18 Number 20 Diameter Distribution Full Sample (N = 139) Non-detections (N = 17) 20 15 10 5 0 0 2 4 6 8 10 12 Diameter [kpc] Number 40 (c) Absolute Magnitudes Full Sample (N = 139) Non-detections (N = 17) 30 20 10 0 -12 -14 207 velocity centroid of the Local Group using the standard correction (Vcor = Vhelio + 300 sin l cos b.). The H i mass MH i is computed from the H i flux integral and distance D: Z 5 2 MH i ¼ 2:36  10 D ðS dvÞc M (a) Apparent Magnitudes 25 Number H i IN STAR-FORMING DWARF GALAXIES -16 -18 Fig. 2.—Histograms showing some of the fundamental optical properties of our sample: (a) apparent B magnitudes; (b) diameters, in kiloparsecs, measured at the 25.0 B mag arcsec
2 level; (c) absolute B magnitudes. In all cases, the lower dashed histogram shows the properties of the 17 nondetected objects. The median of each distribution is indicated. resent physically small systems. Inspection of Table 2 shows that the nondetected galaxies typically have small angular sizes as well, although a few large galaxies are deficient in H i. The majority of the galaxies in the sample are nearby; the median recession velocity is 1710 km s
1, and only 25% of the galaxies are farther than 2500 km s
1 away. This localization of the sample is due to the luminosity cutoff imposed on our survey list, combined with the brightness limits of the various objective-prism surveys employed. Figure 2c illustrates the luminosity distribution. The median B-band absolute magnitude is
16.1. For comparison, the luminosities of the Large and Small Magellanic Clouds are
18.1 and
16.7, respectively (Bothun & Thompson 1988; van den Bergh 2000). The sharp drop at MB =
17 is a direct result of our selection criterion that stipulated that only galaxies fainter in absolute magnitude than this value were to be included. This diagram, along with Figure 2b, confirms and emphasizes the dwarf nature of the sample—these galaxies are true low-luminosity systems. 2.6. Derivation of Physical Quantities Several of the quantities listed in Table 3 were computed from the observational data; their derivation is described in this section. The absolute magnitude MB is derived from the distance D and the apparent magnitude, corrected for Galactic absorption AB. The distance is obtained from the observed redshift, assuming a pure Hubble expansion with H0 = 75 km s
1 Mpc
1. The redshifts are corrected to the (Roberts 1975), where D is in megaparsecs and the flux integral has units of Jy km s
1. The flux integral is corrected for beam-size effects following standard practice (e.g., Thuan & Martin 1981; Staveley-Smith et al. 1992): fc ¼ ½1 þ ðaH i =Þ2 1=2 ½1 þ ðbH i =Þ2 1=2 ; where h is the beam size of the Arecibo telescope (3<3) and aH i and bH i are the major and minor axes of the H i gas (in arcminutes). Here we assume that the H i major-axis diameter is 3.1 times larger than the optical diameter, and that the axial ratio of the H i gas is the same as the optical axial ratio. The value for dH i/d25 comes from comparing our optical sizes with 14 galaxies mapped in H i by Taylor et al. (1995). It should be noted that there is a large variation in the ratio dH i/d25 for these 14 galaxies (observed range of 1.5–7.2 for this sample of 14 BCDs, with a standard deviation of 1.3 and an error in the mean of 0.34), so our adopted value of 3.1 carries with it a large uncertainty. The beamwidth correction factors fc for our sample are fairly small: for 82% of the galaxies, the correction factor is less than 1.35, while it exceeds 2.0 for only four of 122 detected galaxies. The total (dynamical) masses listed in column (11) were derived based on the observed H i velocity widths following the procedure given by Staveley-Smith et al. (1992). The computation makes use of the linear size of the galaxy, which was taken to be dH i = 3.1d25. It should be stressed that determining dynamical masses for dwarf galaxies is fairly uncertain even when high-quality H i maps are available. In particular, it is notoriously difficult to assign reliable inclination values to dwarf galaxies, since their intrinsic shapes are not known and may well vary from object to object. Hence, masses derived using only profile widths should be treated as being representative values only. 3. H i PROPERTIES OF DWARF STAR-FORMING GALAXIES The results of our observations are summarized in Figures 3 and 4, which reveal the H i properties of this sample of star-forming dwarf galaxies. Figure 3 presents a series of histograms showing the distributions of H i velocity widths (Fig. 3a), H i masses (Fig. 3b), and total (dynamical) masses (Fig. 3c) for our sample. The median value for W50 for our galaxies is 88 km s
1, and 75% of the galaxies have widths under 120 km s
1. For comparison, the sample of blue compact galaxies observed by Thuan & Martin (1981) have a median W50 = 97 km s
1, while the median width is 84 km s
1 for those observed by Staveley-Smith et al. (1992). Velocity widths for large spiral galaxies typically fall in the range of 200–600 km s
1 (e.g., Haynes & Giovanelli 1984). It is worth noting that the observed widths do not continue down to arbitrarily low values; the lowest value measured is 33 km s
1 for Mrk 1338. This is in contrast to studies of quiescent, extremely low mass dwarf irregular galaxies (e.g., Carignan, Beaulieu, & Freeman 1990; Lo, Sargent, & Young 1993; Côté 1995; Young & Lo 1997), where many 208 SALZER ET AL. Number 20 (a) HI Velocity Widths Vol. 124 10 (a) 15 9 10 5 8 0 0 50 100 150 200 7 Number 20 (b) HI Masses 6 15 -12 -13 -14 -15 -16 -17 -18 -19 10 1.5 5 (b) Location of Progenitors 1 0 7 8 9 10 .5 Number 20 (c) Total Masses Median 0 15 -.5 10 -1 5 -1.5 0 7 8 9 10 Fig. 3.—Histograms showing the distribution of quantities measured or derived from our H i observations. (a) Velocity widths, measured at 50% of the peak height. The median value of 88 km s
1 is marked. (b) Logarithm of H i masses (MH i), in solar masses. The median value of 3.0  108 M is shown. (c) Logarithm of total (dynamical) masses (Mtot), in solar masses. The median value of 2.4  109 M is indicated. are found to have substantially lower velocity widths. Note that the lack of low velocity width galaxies cannot be a spectral resolution effect, as the majority of the observations were obtained with 2 or 4 km s
1 channel widths. There appears to be a threshold width at roughly 40 km s
1 (or a maximum disk rotational velocity of 20 km s
1) below which actively star-forming dwarf galaxies are not found. This is possibly due to the selection effects present in the objective-prism surveys used to generate our observing sample: these surveys may not be sensitive to extremely low luminosity star-forming galaxies. However, this seems rather unlikely in the case of the UM and CG surveys, which have faint limiting magnitudes and select via line emission, which tends to be stronger (i.e., higher equivalent widths) in lower luminosity systems. This width threshold may represent a real physical limit, indicating the point below which galaxies do not possess a mass concentration sufficient to initiate and sustain a global star formation episode strong enough to put the galaxy into the class of actively star-forming dwarfs of the type detected in the objective-prism surveys used to construct the current sample. The paucity of strong starbursts among extreme dwarfs has also been noted by Boroson, Salzer, & Trotter (1993). Another possibility for the lack of low velocity width systems could be feedback processes (e.g., supernova heating, shocks) from the active star formation going on within the galaxies. The energy injection from the star formation activity may increase the velocity dispersion of the neutral gas up to the observed minimum level, regardless of the mass of the galaxy. -12 -13 -14 -15 -16 -17 -18 -19 Fig. 4.—Comparison between the H i properties of our sample and those of ‘‘ normal ’’ galaxies. (a) Logarithm of the H i mass vs. the absolute B magnitude; (b) logarithm of the H i mass–to–blue light ratio vs. the absolute B magnitude. In both panels, filled circles are the detected galaxies from the current study, the triangles have only H i upper limits, and the open circles with error bars represent the mean values of the plotted quantities for the comparison sample (see text for details). The star-forming galaxies appear to have mean properties very similar to the more quiescent galaxies in the comparison sample, albeit with much larger scatter. However, the similarities in the average properties of the two samples disappear when the starforming galaxies are corrected for the light from the current starburst. The irregular dashed outline in the bottom panel indicates the location of the star-forming dwarfs after a correction is applied to their luminosities (see text). The direction and magnitude of the luminosity correction are indicated by the arrow. The histograms of H i and total mass further reinforce the diminutive nature of the sample galaxies. The most massive objects have less than 10% of the mass of the Milky Way (MMW 8  1011 M), while the least massive are roughly 8000 times lower. In H i, the star-forming dwarfs appear to be fairly gas-rich, with the highest MH i being only a factor of 2 lower than the Milky Way value, while the median value is 1/30 the amount in our Galaxy. The distribution of H i masses in the Thuan & Martin and Staveley-Smith et al. studies is similar. The most extreme low-mass cases (e.g., CG 166, with 3.4  106 M, and UM 538, with 4.9  106 M) have roughly 2000 times less H i gas than the Milky Way. Note that Figure 3b does not include the upper limits for the 17 nondetections. All but one of the upper limits are below the median value of 3.0  108 M, indicating that the sample median is slightly elevated above its true value. Figure 4 shows the logarithms of MH i and the H i mass– to–blue light ratio (MH i/LB) versus blue absolute magnitudes for the sample galaxies. For comparison, in both plots we also show the mean values of these quantities for three large H i data sets: Giovanelli & Haynes (1983) for isolated No. 1, 2002 H i IN STAR-FORMING DWARF GALAXIES field spiral galaxies covering the luminosity range MB =
22 to
17, Fisher & Tully (1975) for the DDO catalog of dwarf and irregular galaxies covering MB =
19 to
13, and extreme dwarf galaxies observed by Sargent & Lo (1986) for MB fainter than
15. These data sets were chosen because of their large sizes and the fact that the H i data were obtained in a uniform manner. We use these data to represent the H i characteristics of ‘‘ normal ’’ galaxies (i.e., the population of less active, relatively quiescent galaxies). The galaxies in our sample overlap primarily with the DDO catalog galaxies, which are mostly Magellanic and dwarf irregular galaxies. The data are presented here as averages within bins 0.5 or 1.0 mag wide. The error bars reflect the standard deviation about the mean for each bin. For our sample of star-forming dwarfs, the nondetections are plotted as upper limits using triangles. In both plots, the star-forming dwarfs are seen to follow very closely the trends established by the ‘‘ normal ’’ galaxies covering the same luminosity range. The mean values for both MH i and MH i/LB for the star-forming dwarf galaxies agree closely with those of the comparison galaxies in each luminosity bin. Two important points are immediately obvious: (1) the scatter in the H i properties of the starforming dwarfs is much larger than that of the ‘‘ normal ’’ galaxies, and (2) in the mean, the star-forming dwarfs have very similar H i properties to ‘‘ normal ’’ galaxies of comparable optical luminosity. We discuss both of these points in more detail below. Figure 4b illustrates the large scatter in the H i properties of the star-forming dwarfs, compared with the ‘‘ normal ’’ galaxies. In the luminosity range
14 to
16, the standard deviation about the mean values of log (MH i/LB) for the star-forming dwarfs is more than double that of the comparison sample. We interpret this as being due to the very heterogeneous nature of the current sample. For example, there are several objects in the sample in which the starburst is hosted by a dwarf elliptical galaxy. These include UM 465, Mrk 324, Mrk 328, and Mrk 900. In all cases, the H i is fairly weak but clearly detected. These objects all have low MH i/LB and are actually gas-poor relative to the other objects. However, they probably are all gas-rich compared with other dwarf elliptical galaxies (see, e.g., Bothun et al. 1985). At the other end of the spectrum are very intense starforming systems such as II Zw 40, Mrk 36 (=CG 82), and UM 439. Although not necessarily extreme in their H i properties, these systems possess some of the most powerful starbursts known (at least in a relative sense). The large range in properties of the sample galaxies, in terms of morphology, physical size, and starburst strength, must be kept in mind when interpreting the results of this H i survey. The scatter seen in Figure 4 can best be understood by appreciating the diversity in the sample of galaxies observed. The similarities in the mean values of the H i properties for the star-forming dwarfs and the comparison sample plotted in Figure 4 are largely illusory. This is due to the fact that the optical characteristics of these galaxies are significantly altered by the occurrence of the global star formation episode that has caused them to be selected for inclusion in the objective-prism surveys used to generate our sample. The starburst will cause each galaxy to increase in optical luminosity by an average of 0.75 mag in the B band (Salzer & Norton 1999). This will move the galaxies to the left in Figure 4a (which plots log MH i vs. MB), and to the left and 209 up in Figure 4b [which plots log (MH i/LB) vs. MB], now placing them significantly above the ‘‘ normal ’’ galaxies of the same luminosity. This effect is illustrated in Figure 4b, where we plot a dashed outline that shows where the galaxies would be located if the luminosity of each galaxy were reduced by 0.75 mag. The galaxies have been shifted to the left in the diagram by 0.75 mag, and upward by 0.3 in MH i/LB. The magnitude and direction of this shift are illustrated by the arrow in the upper left portion of the figure. Of course, this value is simply representative; some galaxies will brighten by more than this (the extreme BCDs), while others will experience a more modest increase. The purpose of the figure is to illustrate the effect of the starburst on the observed properties of these galaxies. With the luminosity correction applied, the dashed outline is effectively plotting the locations of the progenitors of the currently observed starbursts. That is, Figure 4b shows what the host galaxies of the strong starbursts were like before the star formation event took place. All the evolution is assumed to occur in the optical properties, while the H i mass does not decrease significantly as a result of the starburst. Although not true in detail, the amount of mass converted into stars during the starburst phase is on the order of 105–106 M, which is less than 1% of the H i mass for most of the galaxies. This exercise reveals that the host galaxies of BCD-like starbursts are, on average, gas-rich, with many of them located above the region occupied by the comparison sample galaxies in Figure 4. With the luminosity correction applied, the mean MH i/LB of the star-forming dwarfs is a factor of 2 higher than the comparison sample, with 61 of 139 (44%) lying above the 1  error bars plotted for the comparison galaxies. This suggests that the progenitors of at least some star-forming dwarf galaxies are not found in catalogs of ‘‘ normal ’’ dwarf galaxies, since otherwise the upper region of the diagram would be populated by galaxies from both samples. The fact that the high-MH i/LB galaxies in our sample lie above the comparison sample (mostly DDO dwarfs, which tend to be low surface brightness in nature) implies that it is highly unlikely that the typical dwarf galaxy hosting a major starburst will be found among the gas-rich, relatively low surface brightness (LSB) dwarf irregular galaxies. Given this result, where is the progenitor population of the observed star-forming dwarf galaxies? It is often argued that a BCD galaxy will spend most of its evolutionary lifetime in the inactive state, meaning that the space density of inactive BCDs should greatly exceed those of the currently active dwarfs. The fact that at least some of the progenitor galaxies have properties unlike the ‘‘ normal ’’ comparison sample galaxies implies that they are hard to detect. One clue is simply that the host dwarf galaxies of major starbursts are systematically more compact, and have higher central mass densities, than do less active dwarf irregular galaxies. Indeed, many of the objects in the UM and Wasilewski surveys appear quite stellar on the POSS, and it is only the presence of emission lines that gives their extragalactic nature away. While the comparison samples were taken from more traditional optical catalogs, the selection function for the objective-prism surveys tends to select objects that are more extreme and hence tend to be missed in optical surveys (e.g., ultracompact galaxies). Norton & Salzer (2002) discuss in detail the implications of these structural differences between BCDs and dwarf irregular galaxies 210 SALZER ET AL. in the context of the evolutionary history of BCDs. They conclude that galaxies that host BCD-like starbursts are preferentially the most centrally concentrated of the gasrich dwarf galaxies and represent one extreme in the continuum of central surface brightnesses/central mass densities in dwarfs. The data of Staveley-Smith et al. (1992) suggest a trend in the mass-to-light ratio with luminosity of the form MH i/ LB / LB
0:3 . While such a relationship provides a reasonable fit to the ‘‘ normal ’’ galaxies at higher luminosities, no such trend is evident in Figure 4b for either the ‘‘ normal ’’ or the star-forming galaxies in the range of absolute magnitudes covered by the latter. Rather, MH i/LB appears to be essentially constant through the luminosity range covered by the star-forming dwarf galaxies. This appearance is confirmed by fits to the binned data, which show that MH i/LB is independent of LB over the range
12 to
17 for both sets of objects. While the galaxies in our sample tend to have a compact, high surface brightness appearance in the optical, their global H i properties are rather similar to those of LSB dwarf galaxies (see, e.g., Schombert, Pildis, & Eder 1997; O’Neil, Bothun, & Schombert 2000; Eder & Schombert 2000). That is, the total H i masses and velocity widths of the LSB galaxies substantially overlap the values for the star-forming dwarf galaxies when only the dwarf LSB galaxies are considered. On average, the LSB dwarfs tend to have larger values of MH i/LB, as one might expect. However, there are examples of high-MH i/LB star-forming dwarf galaxies as well. The main difference in the H i properties between the LSB and star-forming dwarf galaxies appears to be in the distribution of the gas. For example, both Taylor et al. (1994) and van Zee et al. (1998a) found that actively star-forming galaxies tend to have much stronger central concentrations of gas, while LSB galaxies tend to have relatively flat profiles. The centrally peaked H i distribution is likely to be an important requirement for the production of a global star formation event of the type seen in BCDs. The similarities in the global H i properties of the LSB and star-forming dwarf galaxies should not be taken as evidence that the two types are evolutionarily linked. Rather, they appear to be on opposite ends of the distribution of gas-rich dwarf galaxies (see, e.g., Salzer & Norton 1999). We have already alluded to the variety present in the shapes of the H i profiles. It is instructive to examine these profiles carefully. As is typical of dwarf galaxies, the majority exhibit roughly Gaussian, single-peaked profiles rather than the double-horned profiles typical of spiral disks. A few examples of the latter include CG 13, Was 6, CG 152, CG 187, and Mrk 1384. As pointed out by Skillman (1996), the existence of single-peaked profiles cannot be taken as evidence that these dwarfs are not rotationally supported. The majority of all H i maps of dwarfs show that they are dominated by more or less orderly rotation. The differences in the profile shapes of spirals and dwarfs are due primarily to differences in their gas distributions: spirals tend to have central H i cavities, while the majority of the H i gas is located in regions where the rotation curve is flat. In contrast, the gas in dwarfs is more centrally located, and they tend to have rising rotation curves throughout much of their gaseous disks (e.g., Taylor et al. 1994; van Zee et al. 1998a). Many variations on the Gaussian profile theme are seen in Figure 1. A large proportion show some form of asymme- Vol. 124 try; two very pronounced examples are UM 372 and Was 28. Was 29 and Mrk 756 are interesting cases in which the profiles show steep sides characteristic of double-horned profiles, but in both cases one of the ‘‘ horns ’’ is missing. The profiles of CG 112, Was 52, and II Zw 71 are fairly symmetric, but they show a strong central ‘‘ cusp ’’ of H i emission, perhaps indicating an unusually strong central concentration of gas. II Zw 71 is particularly interesting in this regard. Finally there are the unusual triangular profiles, of which Mrk 416, Haro 38, and Mrk 900 are the best examples. Understanding the details of the profile shapes in individual cases will require H i mapping. One of the original motivations of this survey was to identify examples of BCDs with strong H i and unusual profile shapes that would be good candidates for follow-up mapping with the VLA. Such maps already exist for a number of the galaxies in our sample, most notably in Taylor et al. (1995) and van Zee et al. (1998a). Comparison of our single-dish profiles and the VLA maps of Taylor et al. allows one to better visualize the origins of the irregularities present. For example, the asymmetry in UM 372 is clearly seen in the VLA map to be due to an off-center H i distribution in this galaxy, where the approaching side of the disk contains more gas than the receding side. Similar asymmetries in the profiles of UM 422 and UM 439 can be understood upon inspection of the maps. Another excellent example is UM 456, for which the Arecibo profile shows a strong, fairly symmetric profile and a weaker companion profile offset in velocity by about 150 km s
1. The map of Taylor et al. shows UM 456 to have a very complex H i distribution, indicative of a pair of interacting galaxies. The companion object is clearly visible in the H i map at the velocity seen in our single-dish spectrum. The asymmetries present in our profile of II Zw 40 can also be understood by comparing with the map of van Zee et al. (1998a), although the modest beam size of the Arecibo telescope excludes much of the H i tail that extends far to the southeast from the optical galaxy. Finally we remind the reader that while these galaxies appear to be H i–rich, little is known about their molecular gas content. While molecular gas appears to make up roughly 50% of the gas mass in typical spiral galaxies, it seems to be a lesser constituent in dwarfs. Efforts to observe CO emission from star-forming dwarfs have met with mixed results (Sage et al. 1992; Taylor et al. 1998; Barone et al. 2000). While CO has been convincingly detected in several systems, the overall rate of success at finding molecular emission is low. Somewhat surprisingly, starbursts hosted by dwarf elliptical galaxies are readily detected in CO (Sage et al. 1992; Kobulnicky et al. 1995), while it is rarely seen in those hosted by H i–rich dwarf irregular galaxies. In all cases studied by Sage et al. (1992), the H i mass exceeded the molecular gas mass. However, because of uncertainties in the CO–to–molecular hydrogen conversion factors in lowmetallicity gas, the true contribution of molecular gas to the total gas mass remains quite uncertain in dwarf galaxies. 4. SUMMARY AND CONCLUSIONS The 305 m telescope of the Arecibo Observatory was used to observe 139 dwarf galaxies selected from objective-prism survey lists such as the Markarian, Michigan, Wasilewski, and Case surveys. This represents a complete list of all galaxies in these surveys that could be observed from Arecibo No. 1, 2002 H i IN STAR-FORMING DWARF GALAXIES and were known to have MB >
17.0. A total of 122 galaxies were detected (88%). Our sample is fairly heterogeneous, comprising galaxies with a large range in star formation characteristics, from extreme BCDs such as Mrk 36 and UM 439 to blue post-starburst dwarfs with little star formation activity remaining. It also includes all different morphological types, from gas-rich dwarf irregular to Magellanic irregular to dwarf elliptical galaxies. The one common element in the sample is their presence in one of the aforementioned surveys, which indicates either ongoing or recent star formation activity. The luminosity limit imposed means that our sample of galaxies is truly a dwarf sample. The median absolute magnitude is
16.1 (about the luminosity of the SMC), and the median size is 3.2 kpc. The reader is reminded that all luminosities refer to the current, starburst-enhanced values; the preburst galaxies are typically 0.75 mag less luminous (Salzer & Norton 1999). As would be expected from such a dwarf sample, the observed velocity widths tend to be quite narrow. The median value of W50 is 88 km s
1, and 75% have widths less than 120 km s
1. Few galaxies were observed with widths below 40 km s
1, suggesting that there is a lower limit to how massive a galaxy can be and still host a star formation event of sufficient magnitude to cause it to be selected by the objective-prism surveys used to construct the current sample. Great variety is seen in the H i profiles of these objects. Unusual shapes and asymmetric profiles are the norm rather than the exception. A comparison with existing H i maps provides some insight into the nature of the profile asymmetries and suggests that further H i mapping would be fruitful. The median H i mass is MH i = 3.0  108 M, roughly 1/30 that of the Milky Way, while the overall sample covers 211 a range of a factor of 1300, from 3.4  106 to 4.4  109 M. Although a wide range is seen in the gas content of the sample galaxies, they are on average H i–rich. After correcting the optical luminosities for the enhancement due to the starburst, a large fraction of the star-forming dwarf galaxies exhibit values of MH i/LB > 2, placing them above the range of values seen in samples of ‘‘ normal ’’ galaxies. This suggests to us that the progenitors of at least some of these objects are not found in catalogs of ‘‘ normal ’’ dwarf galaxies and implies that the host galaxies for these major starbursts are typically not found among the population of quiescent dwarf irregular galaxies. Optical imaging studies (e.g., Papaderos et al. 1996a, 1996b; Doublier et al. 1997; Telles, Melnick, & Terlevich 1997; Salzer & Norton 1999; Norton & Salzer 2002) confirm this, showing that the BCDlike starbursts tend to occur in a population of very compact dwarf galaxies. We wish to thank the staff of the Arecibo for their usual excellent assistance during our observing runs for this project. Special thanks goes to Riccardo Giovanelli, for his assistance, access to his GALPAC software package, and numerous discussions about H i in galaxies, and to Angel Vasquez for his golfing tips during our late-night observing sessions. We also thank the anonymous referee, whose many suggestions helped to improve the paper, and Katherine Rhode for her careful reading of the manuscript and valuable comments. Support for this project was provided by NAIC, the Ford Foundation, Wesleyan University, and through a Cottrell College Science Grant from the Research Corporation to J. J. S. The research of J. M. M. is supported by the Jet Propulsion Laboratory, California Institute of Technology, under contract with NASA. REFERENCES Barone, L. T., Heithausen, A., Hüttenmeister, S., Fritz, T., & Klein, U. Meurer, G. R., Staveley-Smith, L., & Killeen, N. E. B. 1998, MNRAS, 300, 2000, MNRAS, 317, 649 705 Boroson, T. A., Salzer, J. J., & Trotter, A. 1993, ApJ, 412, 524 Nilson, P. 1973, Uppsala General Catalogue of Galaxies (Uppsala: Bothun, G. D., Halpern, J. P., Lonsdale, C. J., Impey, C., & Schmitz, M. Uppsala Astron. Obs.) 1989, ApJS, 70, 271 Norton, S. A., & Salzer, J. J. 2002, in preparation Bothun, G. D., Mould, J. R., Wirth, A., & Caldwell, N. 1985, AJ, 90, 697 O’Neil, K., Bothun, G. D., & Schombert, J. 2000, AJ, 119, 136 Bothun, G. D., & Thompson, I. B. 1988, AJ, 96, 877 Papaderos, P., Loose, H.-H., Fricke, K. J., & Thuan, T. X. 1996a, A&A, Bottinelli, L., Chamaraux, P., Gouguenheim, L., & Heidmann, J. 1973, 314, 59 A&A, 29, 217 Papaderos, P., Loose, H.-H., Thuan, T. X., & Fricke, K. J. 1996b, A&AS, Bridle, A. H., Davis, M. M., Fomalont, E. B., & Lequeux, J. 1972, AJ, 77, 120, 207 405 Pesch, P., & Sanduleak, N. 1983, ApJS, 51, 171 (erratum 53, 945) Brinks, E., & Klein, U. 1988, MNRAS, 231, 63P Pesch, P., Stephenson, C. B., & MacConnell, D. J. 1995, ApJS, 98, 41 (erraBurstein, D., & Heiles, C. 1984, ApJS, 54, 33 tum 103, 513 [1996]) Carignan, C., Beaulieu, S., & Freeman, K. C. 1990, AJ, 99, 178 Roberts, M. S. 1975, in Stars and Stellar Systems, Vol. 9, Galaxies and the Côté, S. 1995, Ph.D. thesis, Australia Natl. Univ. Universe, ed. A. Sandage, M. Sandage, & J. Kristian (Chicago: Univ. Doublier, V., Comte, G., Petrosian, A., Surace, C., & Turatto, M. 1997, Chicago Press), 309 A&AS, 124, 405 Sage, L. J., Salzer, J. J., Loose, H.-H., & Henkel, C. 1992, A&A, 265, 19 Eder, J. A., & Schombert, J. M. 2000, ApJS, 131, 47 Salzer, J. J. 1989, ApJ, 347, 152 Fisher, J. R., & Tully, R. B. 1975, A&A, 44, 151 Salzer, J. J., & MacAlpine, G. M. 1988, AJ, 96, 1192 Gerola, H., Seiden, P. E., & Schulman, L. S. 1980, ApJ, 242, 517 Salzer, J. J., MacAlpine, G. M., & Boroson, T. A. 1989a, ApJS, 70, 447 Giovanelli, R., & Haynes, M. P. 1983, AJ, 88, 881 ———. 1989b, ApJS, 70, 479 Gordon, D., & Gottesman, S. T. 1981, AJ, 86, 161 Salzer, J. J., Moody, J. W., Rosenberg, J. L., Gregory, S. A., & Newberry, Haro, G. 1956, Bol. Obs. Tonantzintla Tacubaya, 2, 8 M. V. 1995, AJ, 109, 2376 Haynes, M. P., & Giovanelli, R. 1984, AJ, 89, 758 Salzer, J. J., & Norton, S. A. 1999, in IAU Colloq. 171, The Low Surface Hoffman, G. L., Helou, G., Salpeter, E. E., & Lewis, B. M. 1989, ApJ, 339, Brightness Universe, ed. J. I. Davies, C. Impey, & S. Phillipps (ASP 812 Conf. Ser. 170) (San Francisco: ASP), 253 Kobulnicky, H. A., Dickey, J. M., Sargent, A. I., Hogg, D. E., & Conti, Sargent, W. L. W., & Lo, K. Y. 1986, in Star-forming Dwarf Galaxies and P. S. 1995, AJ, 110, 116 Related Objects, ed. D. Kunth, T. X. Thuan, & J. Trân Thanh Vân (Ed. Lequeux, J., & Viallefond, F. 1980, A&A, 91, 269 Frontières: Gif-sur-Yvette), 253 Lo, K. Y., Sargent, W. L. W., & Young, K. 1993, AJ, 106, 507 Schombert, J. M., Bothun, G. D., Impey, C. D., & Mundy, L. G. 1990, AJ, MacAlpine, G. M., Smith, S. B., & Lewis, D. W. 1977, ApJS, 34, 95 100, 1523 MacAlpine, G. M., & Williams, G. A. 1981, ApJS, 45, 113 Schombert, J. M., Pildis, R. A., & Eder, J. A. 1997, ApJS, 111, 233 Markarian, B. E. 1967, Astrofizika, 3, 55 Searle, L., & Sargent, W. L. W. 1972, ApJ, 173, 25 Markarian, B. E., Lipovetsky, V. A., & Stepanian, D. A. 1981, Astrofizika, Skillman, E. D. 1996, in ASP Conf. Ser. 106, The Minnesota Lectures on 17, 619 (English transl. Astrophysics, 17, 321 [1982]) Extragalactic Neutral Hydrogen, ed. E. D. Skillman (San Francisco: Mazzarella, J. M., & Balzano, V. A. 1986, ApJS, 62, 751 ASP), 208 Meurer, G. R., Carignan, C., Beaulieu, S. F., & Freeman, K. C. 1996, AJ, Smoker, J. V., Davies, R. D., Axon, D. J., & Hummel, E. 2000, A&A, 361, 111, 1551 19 212 SALZER ET AL. Staveley-Smith, L., Davies, R. D., & Kinman, T. D. 1992, MNRAS, 258, 334 Taylor, C. L., Brinks, E., Grashuis, R. M., & Skillman, E. D. 1995, ApJS, 99, 427 (erratum 102, 189 [1996]) Taylor, C. L., Brinks, E., Pogge, R. W., & Skillman, E. D. 1994, AJ, 107, 971 Taylor, C. L., Brinks, E., & Skillman, E. D. 1993, AJ, 105, 128 Taylor, C. L., Kobulnicky, H. A., & Skillman, E. D. 1998, AJ, 116, 2746 Telles, E., Melnick, J., & Terlevich, R. 1997, MNRAS, 288, 78 Thuan, T. X., Lipovetsky, V. A., Martin, J.-M., & Pustilnik, S. A. 1999, A&AS, 139, 1 Thuan, T. X., & Martin, G. E. 1981, ApJ, 247, 823 van den Bergh, S. 2000, PASP, 112, 529 van Zee, L., Salzer, J. J., & Skillman, E. D. 2001, AJ, 122, 121 van Zee, L., Skillman, E. D., & Salzer, J. J. 1998a, AJ, 116, 1186 van Zee, L., Westpfahl, D., Haynes, M. P., & Salzer, J. J. 1998b, AJ, 115, 1000 Viallefond, F., Lequeux, J., & Comte, G. 1987, in Starbursts and Galaxy Evolution, ed. T. Thuan, T. Montmerle, & J. Trân Thanh Vân (Ed. Frontières: Gif-sur-Yvette), 139 Viallefond, F., & Thuan, T. X. 1983, ApJ, 269, 444 Wasilewski, A. J. 1983, ApJ, 272, 68 Young, L. M., & Lo, K. Y. 1997, ApJ, 490, 710 Zwicky, F. 1971, Catalogue of Selected Compact Galaxies and of Posteruptive Galaxies (Gümligen: F. Zwicky) Zwicky, F., Herzog, E., Kowal, C. T., Wild, P., & Karpowicz, M. 1961– 1968, Catalogue of Galaxies and of Clusters of Galaxies (Pasadena: Caltech)