THE OBSERVED PROPERTIES OF DWARF GALAXIES IN AND AROUND THE LOCAL GROUP

There has been a veritable explosion of data, discoveries, and realizations in the past decade relating to the broad research area of the structure and content of nearby galaxies, particularly those in which it is possible to resolve the individual stars that contribute to the galaxy's luminosity, and to which this article exclusively refers. By necessity, the objects of the attention of this research are the Milky Way (MW), its satellites, and their neighbors within the so-called Local Group, but it is ever expanding to include the far richer sample of objects that fall within the loosely defined part of the universe referred to only as the Local Volume.

The rapid growth in interest in this field can be traced to two distinct developments that have occurred in tandem. The first is entirely observational: nearby galaxies subtend large areas on the sky and contain a high density of individually rather faint stellar sources. Thus, the advent of wide-field, digital, high-resolution, multiplexing capabilities on large telescopes (and the computational facilities to handle the resulting data-flow) has had a profound effect (and will continue to do so). The second is the result of an increased understanding of the nature of our immediate cosmic environment, facilitated in no small part by computational tools (hardware and software) that have allowed for a higher-resolution examination of the consequences of the prevailing cosmological paradigm. The discoveries and realizations that these advances have enabled are significant: the hierarchical nature of galaxy formation and the realization that some of the relics of that process may still be identifiable at z = 0; the quest for a better understanding of dark matter and the realization that the smallest galaxies nearest to us may be the darkest laboratories in the universe; the chemical evolution of the universe and the realization that the nucleosynthetic imprints of the earliest generations of star formation may still be found in the Galaxy and its satellites. These are but a few examples. A new phrase has developed to collectively describe the aforementioned research and its motivations, and which contrasts it with its higher-redshift cousins: "near-field cosmology."

The phrase "near-field cosmology" is relatively recent and dates from the Annual Review article of Freeman & Bland-Hawthorn (2002). As such, it is sometimes considered a new field within astrophysics as a whole, although this is grossly unfair to its early exponents. Indeed, what could be argued as two of the most important papers in astrophysics—those of Eggen et al. (1962) and Searle & Zinn (1978)—are quintessential examples of observational near-field cosmology. The former took a well-defined sample of stars for which high-quality spectroscopic and photometric data existed and from that motivated a collapse model for the formation of the Galaxy. The latter used complexities within the stellar populations of the Galactic globular cluster system to motivate a scenario where the Galaxy (or part of it) is built through the accretion of smaller stellar systems. The principle ideas from both papers continue to form critical aspects of modern galaxy formation theory.

Many more galaxies are now able to have their resolved stellar content examined than was possible at the time of these early studies. The content and structure of the Andromeda Galaxy (M31) provide an excellent comparison (and, increasingly, contrast) to our own Galaxy (e.g., Courteau et al. 2011; Huxor et al. 2011; Watkins et al. 2010; McConnachie et al. 2009; Yin et al. 2009; Ibata et al. 2007; Chapman et al. 2007; Kalirai et al. 2006; Koch et al. 2008, and references therein). Even excluding this prominent astronomical target, however, there are still nearly a hundred or so galaxies for which their individual stars can be brought into focus by astronomers wanting to unearth fossil signatures of galactic formation.

The overwhelming majority of nearby galaxies—indeed, the overwhelming majority of galaxies in the entire universe—are dwarf galaxies. Exactly how low a luminosity or how small a mass is required before a galaxy is deemed meager enough to warrant this popular designation is generally not well defined or accepted. Kormendy (1985) famously showed that when surface brightness is plotted against luminosity for elliptical galaxies, two relations are apparently formed, one for "giant" galaxies and one for "dwarf" galaxies. This has usually been interpreted as evidence that these systems should be separated into two distinct groups (in terms of different formation mechanisms). Some more recent studies, on the other hand, suggest that there is actually a continuous sequence connecting the dwarf and giant regimes, and that the change in the form of the scaling relations with luminosity does not necessitate fundamental differences between low- and high-luminosity systems (e.g., Graham & Guzmán 2003, and references therein). With these considerations in mind, and for the purpose of this article, I consider dwarfs to have absolute visual magnitudes fainter than M_V ∼ −18 (also adopted by Grebel et al. 2003). In what follows, this places all nearby galaxies but the MW, the Large Magellanic Cloud, M31, M33, NGC 55, and NGC 300 in the dwarf category.

The closest galaxies to the MW are its own satellites, the most prominent of which are the Large and Small Magellanic Clouds (LMC/SMC), both of which are naked-eye objects. The earliest reference to the Magellanic Clouds that survives is by the Persian astronomer Abd al-Rahman al-Sufi in his 964 work, Book of Fixed Stars (which also includes the earliest recorded reference to the Andromeda Galaxy). A millennium or so later, Shapley (1938a) discovered "a large rich cluster with remarkable characteristics" in the constellation of Sculptor. He speculated that it was

• ...a super-cluster of the globular type and of galactic dimensions; or a symmetrical Magellanic Cloud devoid of its characteristic bright stars, clusters, and luminous diffuse nebulosity; or a nearby spheroidal galaxy, highly resolved, and of abnormally low surface brightness. These phrases are merely different ways of describing the same thing and of pointing out the uniqueness of the object.

Over seventy years later, a large body of research still devotes itself to understanding the detailed properties and structural, chemical, and dynamical characteristics of this class of stellar system. The prototype described above is referred to as the Sculptor dwarf spheroidal (dSph) galaxy.

Excellent discussions of the observational properties of—and our astrophysical understanding of—the nearest (Local Group) dwarf galaxies can be found in a large number of dedicated review articles (e.g., Gallagher & Wyse 1994; Grebel 1997; Mateo 1998; van den Bergh 2000; Geisler et al. 2007; Tolstoy et al. 2009). The compilations of observational data presented in numerous works by S. van den Bergh, in particular van den Bergh (2000), and the Annual Review article by Mateo (1998) are particularly valuable resources for the astronomical community. Since their original publication, however, there have been notable advances (in part a product of the observational and theoretical factors described earlier). Not least of these is the more than doubling of the number of known galaxies in the Local Group. In the surroundings of the Local Group too, there have been significant new discoveries, and techniques and observations that used to be applicable only to the MW satellites or nearby Local Group galaxies can now be applied to these more distant relatives.

The ability to reach the resolved populations of ever more distant galaxies is important not just for the improvement in statistics that it offers, but also in terms of the increasing range of environments that it allows us to study. The overwhelming amount of detailed information regarding the stellar population properties of dwarf galaxies comes from studies of the MW satellite systems. Given the sensitivity of low-mass systems to both internal (e.g., feedback through star formation) and external (e.g., ram and tidal stripping) environmental mechanisms, this might be expected to bias our understanding toward conditions that hold only for the MW environment. Within the very nearby universe, however, there are a large number of environments able to be probed, from the surroundings of the MW and M31, to the more isolated, outlying members of the Local Group, to the very isolated dwarf galaxies that surround the Local Group, to the numerous nearby groups that populate and form our immediate neighborhood. A large number of dedicated surveys in the past few years are systematically examining galaxies in the Local Volume at a range of wavelengths, and which include among their targets some of the more distant galaxies in this sample. These surveys are obtaining precision insights into local galaxies that naturally compliment lower-resolution studies of their more numerous, but more distant, cousins (e.g., SINGS, Kennicutt et al. 2003; the Spitzer LVL, Dale et al. 2009; THINGS, Walter et al. 2008; FIGGS, Begum et al. 2008b; HST ANGRRR/ANGST, Dalcanton et al. 2009; the GALEX survey of galaxies in the Local Volume, Lee et al. 2011; the Hα survey of Kennicutt et al. 2008; and many others).

I have alluded to several motivations for pursuing studies of nearby galaxies, be they related to cosmic chemical processing, dynamical evolution, galactic formation mechanisms, environment, near-field cosmology, or a wealth of other research areas on which much can be written. It is not the purpose of this article to contribute to those discussions here; the recent Annual Review article by Tolstoy et al. (2009) provides an excellent, current overview of the status of research into many of these topics using Local Group galaxies. Rather, this article attempts to collate, homogenize, and critically review our observational understanding of aspects of the nearby populations of dwarf galaxies, with particular emphasis on the origins and uncertainties of some key observable parameters. I will discuss science topics only when such discussions relate directly to observational measurements, and those measurements, in my opinion, provide grounds for caution, or when such discussion provides a context for highlighting observational relationships between parameters. This article primarily focuses on the stellar properties of the galaxies (and derived properties, such as the implied dark matter content) and cites observations conducted mostly at ultraviolet–optical–near-infrared wavelengths, with less discussion on their gaseous and dust content. With the intense efforts currently underway to provide a more full characterization of these systems, I expect that some numbers provided in this article will be out of date by the time it is published. As such, tables presented herein will be continually updated to provide a convenient, online library of dwarf galaxy parameters for general reference and use by the community.

The layout of the article is as follows. Section 2 describes the selection of the sample and gives a general summary of my methodology and reasoning in the construction of the data set. Section 3 reviews the discovery space of the galaxies and discusses general issues relating to distances, velocities, and the membership (or otherwise) of larger-scale groups/sub-groups. Section 4 reviews the observed photometric and structural characteristics of the sample and discusses aspects of the scaling relations that they define. Section 5 reviews their masses (stellar, H i, and dynamical) and internal kinematics and discusses issues related to morphological classifications and our dynamical understanding of the sample. Section 6 presents a compilation of available mean stellar metallicities and discusses associated observational trends. Section 7 concludes and summarizes.

The sample of galaxies discussed in this article consists of all known galaxies with distances determined from measurements of resolved stellar populations (usually Cepheids, RR Lyrae, tip of the red giant branch, TRGB, but also including horizontal branch level and main-sequence fitting; see Tammann et al. 2008 for a thorough discussion of the former three methods and references) that place them within 3 Mpc of the Sun. Consequently, this sample contains the satellite systems of two major galaxies (the MW and M31), the quasi-isolated, outlying members of the Local Group, and the nearby isolated dwarf galaxies that do not clearly belong to any major galaxy grouping. It is therefore expected that this sample may be valuable (as, indeed, various individual members and sub-sets have already proven valuable) for examining the role played by environment in dwarf galaxy evolution, as discussed in Section 1. There is also a practical motivation for the choice of distance limit, namely, that a much larger threshold will start to select members of nearby groups (the closest of which is the Maffei group at ∼3 Mpc, although its awkward location at low Galactic latitudes prevents accurate distances from being determined for its members; see Fingerhut et al. 2007) and would expand the scope of this article considerably.

Of course, there are no solid boundaries between galaxy groups, filaments, and the field, and some of the galaxies discussed herein are likely dynamically associated with neighboring structures like the Sculptor (e.g., Karachentsev et al. 2003a) or IC 342 (e.g., Karachentsev et al. 2003b) groups, for example. Nevertheless, at the time of writing, the adoption of these selection criteria identifies exactly 100 galaxies (not including the MW and M31), of which 73 are definite or very likely members of the Local Group. It is likely that our basic knowledge of all of these galaxies and their resolved stellar content can be improved dramatically over the coming years with existing observational capabilities, and the more distant galaxies provide stepping stones into the Local Volume for future resolved stellar population studies enabled by next-generation facilities like the 30 m class telescopes. It is with an eye to the latter considerations that the distance is not limited to the zero-velocity surface of the Local Group (R_LG).

The ensuing discussion will focus on dwarf galaxies. Of the 102 galaxies that are listed in the subsequent tables, virtually no discussion will be given to the MW and M31, and numbers relating to M33, NGC 55, and NGC 300 are included for completeness only. The same caveat applies to the Magellanic Clouds, since here the research body and available data are so extensive and far exceed those that exists for most other dwarf-like systems that the reader is referred to the many review articles, books, and conferences that deal specifically with these bodies (e.g., Westerlund 1997; van Loon & Oliveira 2009, and references therein).

In compiling this catalog of galaxies, and in addition to the papers cited in each table, I have made extensive use of the NASA Extragalactic Database, the HyperLeda database (Paturel et al. 2003), Mateo (1998), van den Bergh (2000), and Karachentsev et al. (2004). The requirement that all galaxies have distances based on resolved stellar populations leads to the exclusion of a few galaxies that have low velocities that could potentially place them within 3 Mpc, but which lack any direct distance measurement (e.g., see Table 1 of Karachentsev et al. 2004). The low-latitude galaxy Dwingeloo 1 is also excluded by this criterion, since its distance of ∼2.8 Mpc is based on a Tully–Fisher estimate (Karachentsev et al. 2003b; this galaxy is very likely a member of the Maffei group). The (uncertain) TRGB measurement of MCG9-20-1 by Dalcanton et al. (2009) places it at a distance of ∼1.6 Mpc, although its radial velocity of v_☉ ∼ 954 km s⁻¹ is very high if it is really this close, and the photometric data now seem to favor a larger distance (K. Gilbert & J. Dalcanton 2011, private communication), so it too has been excluded from the list. Finally, there are a few galaxies for which it is unclear if they lie closer or farther than 3 Mpc (e.g., UGCA 92, D = 3.01 ± 0.24 Mpc; Karachentsev et al. 2006). I exclude such galaxies, even if the quoted uncertainties would bring them within range of the distance cut. Of course, if future estimates should clearly indicate a closer distance, then the online version of the tables will be updated as appropriate.

In compiling these references, I have favored, where possible, papers that are based on analysis of the resolved stellar populations. In the interest of homogeneity and in an (ill-fated) attempt to minimize the inevitable systematic differences between measurements for different galaxies, I also generally try to limit the number of distinct publications, studies, and/or methodologies that contribute to the overall data set, favoring large surveys of multiple galaxies in preference to studies of individual galaxies. However, if there appear to be significant discrepancies in the literature over observed values, then these differences are indicated either in the footnotes and/or in the text. Infrequently, I have estimated uncertainties where none were given in the original publications.

My primary intent in compiling this data set is to provide useful information on the population of (dwarf) galaxies in the nearby universe. I have tried to ensure that the references collected herein—while not complete in any way—are generally recent enough and relevant enough that I hope they will be able to provide the reader with a starting point from which most of the germane literature can be traced.

Table 1 lists basic information for all nearby galaxies that satisfy the selection criteria discussed in the previous section.

Table 1. Basic Information

(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
Galaxy	Other Names			R.A.	Decl.	Original Publication	Comments
				J2000	J2000
The MW sub-group (in order of distance from the MW)
The Galaxy	The MW	G	S(B)bc	17h45m400	−29d00m28s	...	Position refers to Sgr A*
Canis Major		G	????	07h12m350	−27d40m00s	Martin et al. (2004a)	MW disk substructure?
Sagittarius dSph		G	dSph	18h55m195	−30d32m43s	Ibata et al. (1994)	Tidally disrupting
Segue (I)		G	dSph	10h07m040	+16d04m55s	Belokurov et al. (2007)
Ursa Major II		G	dSph	08h51m300	+63d07m48s	Zucker et al. (2006a)
Bootes II		G	dSph	13h58m000	+12d51m00s	Walsh et al. (2007)
Segue II		G	dSph	02h19m160	+20d10m31s	Belokurov et al. (2009)
Willman 1	SDSS J1049+5103	G	dSph	10h49m210	+51d03m00s	Willman et al. (2005a)	Cluster?
Coma Berenices		G	dSph	12h26m590	+23d54m15s	Belokurov et al. (2007)
Bootes III		G	dSph?	13h57m120	+26d48m00s	Grillmair (2009)	Very diffuse. Tidal remnant?
LMC	Nubecula Major	G	Irr	05h23m345	−69d45m22s	...
SMC	Nubecula Minor	G	dIrr	00h52m448	−72d49m43s	...
	NGC 292
Bootes (I)		G	dSph	14h00m060	+14d30m00s	Belokurov et al. (2006)
Draco	UGC 10822	G	dSph	17h20m124	+57d54m55s	Wilson (1955)
	DDO 208
Ursa Minor	UGC 9749	G	dSph	15h09m085	+67d13m21s	Wilson (1955)
	DDO 199
Sculptor		G	dSph	01h00m094	−33d42m33s	Shapley (1938a)	The prototypical dSph
Sextans (I)		G	dSph	10h13m030	−01d36m53s	Irwin et al. (1990)
Ursa Major (I)		G	dSph	10h34m528	+51d55m12s	Willman et al. (2005b)
Carina		G	dSph	06h41m367	−50d57m58s	Cannon et al. (1977)
Hercules		G	dSph	16h31m020	+12d47m30s	Belokurov et al. (2007)	Tidally disrupting? Remnant? Cluster?
Fornax		G	dSph	02h39m593	−34d26m57s	Shapley (1938b)
Leo IV		G	dSph	11h32m570	−00d32m00s	Belokurov et al. (2007)	Binary w/Leo V?
Canes Venatici II	SDSS J1257+3419	G	dSph	12h57m100	+34d19m15s	Sakamoto & Hasegawa (2006)
						Belokurov et al. (2007)
Leo V		G	dSph	11h31m096	+02d13m12s	Belokurov et al. (2008)	Cluster? Binary w/Leo IV?
Pisces II		G	dSph	22h58m310	+05d57m09s	Belokurov et al. (2010)	Awaiting spectr. confirmation
Canes Venatici (I)		G	dSph	13h28m035	+33d33m21s	Zucker et al. (2006b)
Leo II	Leo B	G	dSph	11h13m288	+22d09m06s	Harrington & Wilson (1950)
	UGC 6253
	DDO 93
Leo I	UGC 5470	G/L	dSph	10h08m281	+12d18m23s	Harrington & Wilson (1950)
	DDO 74
	Regulus Dwarf
The M31 sub-group (in order of distance from M31)
Andromeda	M31	A	Sb	00h42m443	+41d16m09s	...
	NGC 224
	UGC 454
M32	NGC 221	A	cE	00h42m418	+40d51m55s	Legentil (1755)
	UGC 452
Andromeda IX		A	dSph	00h52m530	+43d11m45s	Zucker et al. (2004)
NGC 205	M110	A	dE/dSph	00h40m221	+41d41m07s	Messier (1798)
	UGC 426
Andromeda XVII		A	dSph	00h37m070	+44d19m20s	Irwin et al. (2008)
Andromeda I		A	dSph	00h45m398	+38d02m28s	van den Bergh (1972)
Andromeda XXVII		A	dSph	00h37m271	+45d23m13s	Richardson et al. (2011)	Tidally disrupting? Remnant?
Andromeda III		A	dSph	00h35m338	+36d29m52s	van den Bergh (1972)
Andromeda XXV		A	dSph	00h30m089	+46d51m07s	Richardson et al. (2011)
Andromeda XXVI		A	dSph	00h23m456	+47d54m58s	Richardson et al. (2011)
Andromeda XI		A	dSph	00h46m200	+33d48m05s	Martin et al. (2006)
Andromeda V		A	dSph	01h10m171	+47d37m41s	Armandroff et al. (1998)
Andromeda X		A	dSph	01h06m337	+44d48m16s	Zucker et al. (2007)
Andromeda XXIII		A	dSph	01h29m218	+38d43m08s	Richardson et al. (2011)
Andromeda XX		A	dSph	00h07m307	+35d07m56s	McConnachie et al. (2008)
Andromeda XII		A/L	dSph	00h47m270	+34d22m29s	Martin et al. (2006)	Unbound to M31?
NGC 147	UGC 326	A	dE/dSph	00h33m121	+48d30m32s	Herschel (1833)	Binary w/NGC 185? Tidal stream
	DDO 3
Andromeda XXI		A	dSph	23h54m477	+42d28m15s	Martin et al. (2009)
Andromeda XIV		A/L	dSph	00h51m350	+29d41m49s	Majewski et al. (2007)	Unbound to M31? In Pisces
Andromeda XV		A	dSph	01h14m187	+38d07m03s	Ibata et al. (2007)
Andromeda XIII		A	dSph	00h51m510	+33d00m16s	Martin et al. (2006)	In Pisces
Andromeda II		A	dSph	01h16m298	+33d25m09s	van den Bergh (1972)	In Pisces
NGC 185	UGC 396	A	dE/dSph	00h38m580	+48d20m15s	Herschel (1789)	Binary w/NGC 147?
Andromeda XXIX		A	dSph	23h58m556	+30d45m20s	Bell et al. (2011)	In Pegasus
Andromeda XIX		A	dSph	00h19m321	+35d02m37s	McConnachie et al. (2008)
Triangulum	M33	A	Sc	01h33m509	+30d39m37s	a
	NGC 598
	UGC 1117
Andromeda XXIV		A	dSph	01h18m300	+46d21m58s	Richardson et al. (2011)
Andromeda VII	Casseopia dSph	A	dSph	23h26m317	+50d40m33s	Karachentsev & Karachentseva (1999)
Andromeda XXII		A	dSph	01h27m400	+28d05m25s	Martin et al. (2009)	M33 satellite? In Pisces
IC 10	UGC 192	A	dIrr	00h20m173	+59d18m14s	Swift (1888)
LGS 3 (Local Group	Pisces (I)	A	dIrr/dSph	01h03m550	+21d53m06s	Karachentseva (1976)
Suspect 3)						Kowal et al. (1978)
Andromeda VI	Pegasus dSph	A	dSph	23h51m463	+24d34m57s	Karachentsev & Karachentseva (1999)
						Armandroff et al. (1999)
Andromeda XVI		A/L	dSph	00h59m298	+32d22m36s	Ibata et al. (2007)	In Pisces
The rest of the Local Group and its neighbors (in order of distance from the barycenter of the Local Group)
Andromeda XXVIII		A/L	dSph?	22h32m412	+31d12m58s	Slater et al. (2011)	In Pegasus
IC 1613	DDO 8	L	dIrr	01h04m478	+02d07m04s	Wolf (1906)
	UGC 668
Phoenix		L/G	dIrr/dSph	01h51m063	−44d26m41s	Schuster & West (1976)b
NGC 6822	IC 4895	L/G	dIrr	19h44m566	−14d47m21s	Barnard (1884)	Polar ring morph.
	DDO 209
	Barnard's Galaxy
Cetus		L	dSph	00h26m110	−11d02m40s	Whiting et al. (1999)	Isolated dSph
Pegasus dIrr	UGC 12613	L/A	dIrr/dSph	23h28m363	+14d44m35s	A. G. Wilson. Reported in Holmberg (1958)
	DDO 216
Leo T		L/G	dIrr/dSph	09h34m534	+17d03m05s	Irwin et al. (2007)
WLM (Wolf-Lundmark-	UGCA 444	L	dIrr	00h01m582	−15d27m39s	Wolf (1910)	Defines RLG
-Melotte)	DDO 221					Melotte (1926), including K. E. Lundmark
Leo A	Leo III	L	dIrr	09h59m265	+30d44m47s	Zwicky (1942)	Defines RLG
	UGC 5364
	DDO 69
Andromeda XVIII		L	dSph	00h02m145	+45d05m20s	McConnachie et al. (2008)	Isolated dSph
Aquarius	DDO 210	L	dIrr/dSph	20h46m518	−12d50m53s	van den Bergh (1959)	Defines RLG
Tucana		L	dSph	22h41m496	−64d25m10s	Lavery (1990)c	Isolated dSph; defines RLG
Sagittarius dIrr	UKS 1927-177	L	dIrr	19h29m590	−17d40m41s	Cesarsky et al. (1977)	Defines RLG
						Longmore et al. (1978)
UGC 4879	VV 124	L	dIrr/dSph	09h16m022	+52d50m24s	Kopylov et al. (2008)d	Defines RLG
NGC 3109	DDO 236	N	dIrr	10h03m069	−26d09m35s	Herschel (1847)	Loose NGC 3109 subgroup. Interacting with Antlia?
	UGCA 194
Sextans B	UGC 5373	N	dIrr	10h00m001	+05d19m56s	A. G. Wilson? See Holmberg (1958)e	Loose NGC 3109 subgroup
	DDO 70
Antlia		N	dIrr	10h04m041	−27d19m52s	Whiting et al. (1997)f	Loose NGC 3109 subgroup. Interacting with NGC 3109?
Sextans A	UGCA 205	N	dIrr	10h11m008	−04d41m34s	Zwicky (1942)	Loose NGC 3109 sub-group
	DDO 75
HIZSS 3(A)		N	(d)Irr?	07h00m293	−04d12m30s	Henning et al. (2000)g	Zone of obscuration. Binary with B?
HIZSS 3B		N	(d)Irr?	07h00m293	−04d12m30s	Henning et al. (2000)h	Zone of obscuration. Binary with A?
KKR 25		N	dIrr/dSph	16h13m480	+54d22m16s	Karachentsev et al. (2001b)
ESO 410- G 005	UKS 0013-324	N	dIrr/dSph	00h15m316	−32d10m48s	Lauberts (1982)	NGC 55 sub-group, forming "bridge" to Sculptor?
						Longmore et al. (1982)
NGC 55		N	Irr	00h14m536	−39d11m48s	Dunlop (1828)	NGC 55 sub-group, forming "bridge" to Sculptor?
ESO 294- G 010		N	dIrr/dSph	00h26m334	−41d51m19s	Lauberts (1982)	NGC 55 sub-group, forming "bridge" to Sculptor?
NGC 300		N	Sc	00h54m535	−37d41m04s	Dunlop (1828)	NGC 55 sub-group, forming "bridge" to Sculptor?
IC 5152		N	dIrr	22h02m415	−51d17m47s	D. Stewart. Reported in Pickering (1899)
KKH 98		N	dIrr	23h45m340	+38d43m04s	Karachentsev et al. (2001a)
UKS 2323-326	UGCA 438	N	dIrr	23h26m275	−32d23m20s	Longmore et al. (1978)	NGC 55 sub-group, forming "bridge" to Sculptor?
KKR 3	KK 230	N	dIrr	14h07m105	+35d03m37s	Karachentseva & Karachentsev (1998)i	Member of the Canes Venatici I cloud
GR 8	UGC 8091	N	dIrr	12h58m404	+14d13m03s	Reaves (1956)j	Gibson Reaves gave this galaxy his initials
	VV 558
	DDO 155
UGC 9128	DDO 187	N	dIrr	14h15m565	+23d03m19s	van den Bergh (1959)
UGC 8508		N	dIrr	13h30m444	+54d54m36s	Vorontsov-Vel'Yaminov & Krasnogorskaya (1962)
IC 3104	ESO 020- G 004	N	dIrr	12h18m460	−79d43m34s	D. Stewart. Reported in Pickering (1908)
	UKS 1215-794
DDO 125	UGC 7577	N	dIrr	12h27m409	+43d29m44s	van den Bergh (1959)
UGCA 86		N	dIrr	03h59m483	+67d08m19s	Nilson (1974)	Companion of IC 342
DDO 99	UGC 6817	N	dIrr	11h50m530	+38d52m49s	van den Bergh (1959)
IC 4662	ESO 102- G 014	N	dIrr	17h47m088	−64d38m30s	R. Innes. Reported in Lunt (1902)
DDO 190	UGC 9240	N	dIrr	14h24m434	+44d31m33s	van den Bergh (1959)
KKH 86		N	dIrr	13h54m335	+04d14m35s	Karachentsev et al. (2001a)
NGC 4163	UGC 7199	N	dIrr	12h12m091	+36d10m09s	Herschel (1789)
	NGC 4167k
DDO 113	UGCA 276	N	dIrr	12h14m579	+36d13m08s	van den Bergh (1959)
	KDG 90

Notes. ^aDiscovery credited to G. B. Hodierna (see Steinicke 2010). ^bOriginally thought to be a globular cluster. Canterna & Flower (1977) subsequently identified its galactic nature. ^cThis author first suggested that this object is a member of the Local Group, but it had previously been cataloged by Corwin et al. (1985). ^dThese authors calculate a new distance to UGC 4879 and show that it is on the periphery of the Local Group. While included in the "Atlas and Catalog of Interacting Galaxies" by Vorontsov-Velyaminov (1959), earlier references to this object have not been found. ^eHolmberg (1958) credits Wilson (1955) with discovery, but this object is not listed in this manuscript. ^fThese authors rediscovered this galaxy and showed that it was in the periphery of the Local Group, but it had earlier been cataloged by Corwin et al. (1985), Feitzinger & Galinski (1985), and Arp & Madore (1987). ^gHIZSS 3 resolved into two sources by Begum et al. (2005). Position corresponds to the "HIZSS3 system." ^hHIZSS 3 resolved into two sources by Begum et al. (2005). Position corresponds to the "HIZSS3 system." ⁱSee also Karachentseva et al. (1999). ^jDiscovered on inspection of photographic plates from a survey discussed in Shane (1947). C. D. Shane speculated that some of the nebulae that were visible could be Local Group dwarf galaxies. ^kThe original coordinates of NGC 4167 are coincident with the coordinates of NGC 4163, but Sulentic & Tifft (1973) note that NGC 4167 is "non-existent."

Download table as:  ASCIITypeset images: 1 2 3 4

Column 1. Galaxy name.

Column 2. Common alternative names.

Column 3. Indicator whether they are associated with MW [G], M31 [A], the Local Group [L], or are nearby neighbors [N].

Column 4. Morphological (Hubble) type. The distinction between dwarf elliptical (dE) and dwarf spheroidal (dSph) is based solely on luminosity and is therefore somewhat arbitrary. Objects fainter than M_V ∼ −18 are given a "d" (dwarf) prefix and are again somewhat arbitrary.

Columns 5 and 6. Celestial coordinates (J2000).

Column 7. Original publication. For more details relating to the discoveries of the NGC/IC galaxies, the reader is referred to Steinicke (2010).¹

Column 8. Comments.

Table 2 lists position and velocity information for the galaxy sample.

Column 1. Galaxy name.

Columns 2 and 3. Galactic coordinates (l, b).

Column 4. Foreground extinction, E(B − V): these correspond to values derived by Schlegel et al. (1998) from an all-sky COBE/DIRBE and IRAS/ISSA 100 μm map, at the coordinates of the centroid of each galaxy. While Schlegel et al. (1998) estimate that values for E(B − V) are generally accurate to 16%, a major (and sometimes overlooked, including by this author!) caveat for their use in the direction of nearby (large) galaxies is that the brightest galaxies at IRAS 60 μm² were excised from these maps and replaced by the "most likely value" of the underlying 100 μm emission (see Section 4.2 of Schlegel et al. 1998). The resulting values are not, therefore, directly observed, and any variation across the area occupied by the galaxy is not traced by these maps. For this compilation, the affected galaxies are NGC 6822, IC 1613, M33, NGC 205, NGC 3109, NGC 55, and NGC 300. The LMC, SMC, and M31 (and by necessity also M32) were not excised.

Columns 5 and 6. Distance modulus, heliocentric distance: all distance moduli are based on resolved stellar population analysis (generally Cepheid, RR Lyrae, or TRGB measurements, but also horizontal branch level and main-sequence fitting). Note that many TRGB estimates do not include the formal uncertainty in the absolute magnitude of the TRGB (Bellazzini et al. 2001, 2004a; Rizzi et al. 2007).

Column 7. Heliocentric velocity: for most Local Group galaxies, the velocity corresponds to the optical velocity of the galaxy. For some of the more distant galaxies, velocities are measured from H i.

Columns 8–10. Distance–velocity pairs for the Galactocentric (MW), M31, and Local Group frames of reference: these are calculated by first assuming that the barycenter of the Local Group is located at the mid point of the vector connecting the MW and M31 (for an adopted M31 distance modulus of (m − M)₀ = 24.47; McConnachie et al. 2005). The latter galaxy is more luminous (e.g., Hammer et al. 2007, and references therein), but dynamical mass estimates for both of these galaxies have produced conflicting results regarding which is more massive (e.g., Little & Tremaine 1987; Kochanek 1996; Wilkinson & Evans 1999; Evans & Wilkinson 2000; Evans et al. 2000; Klypin et al. 2002; Ibata et al. 2004; Geehan et al. 2006; Watkins et al. 2010; McMillan 2011), a result nearly exclusively due to the lack of a sufficient number of dynamical tracers at large radii. Given this uncertainty, I conclude that assuming their masses to be roughly the same is reasonable, simple, and convenient. Velocities are converted to the Local Group frame using the solar apex derived by Karachentsev & Makarov (1996) and subtracting the component of the MW's velocity in the direction of each dwarf. M31-centric velocities are obtained by subtracting the component of the Local Group centric velocity of M31 in the direction of each dwarf. Galactocentric distances and velocities are calculated in the usual way, for an adopted Galactic rotation velocity at the Sun of 220 km s⁻¹ at a radius of 8.5 kpc from the Galactic center.

Column 11. References.

Figure 1 shows Aitoff projections of the Galactic coordinates of all galaxies in the sample. The top panel shows only those objects identified as definite or likely MW satellites, the middle panel shows definite or likely M31 sub-group members (blue points) and quasi-isolated Local Group galaxies (green points), and the bottom panel shows galaxies surrounding the Local Group within a distance of 3 Mpc. Some confirmed members of all galaxy groups within 5 Mpc are also shown in the bottom panel (gray points) to highlight the location of the Local Group and its neighbors with respect to these nearby structures (Karachentsev 2005). In this respect, it is important to highlight the work of I. Karachentsev, B. Tully, and their colleagues in both the identification of nearby galaxies and groups and the determination of accurate distances to many of our neighbors in order to understand the structure of the Local Volume (e.g., see Karachentsev & Tikhonov 1994; Karachentsev et al. 2003a, 2004; Tully 1987; Tully et al. 2006, 2009).

Zoom In Zoom Out Reset image size

3.1. Membership of the MW Sub-group

Figure 2 shows Galactocentric velocity versus distance for all galaxies within 600 kpc for which necessary data exist. Here, the Galactocentric velocity listed in Table 2 has been multiplied by a factor of $\sqrt{3}$ to provide a crude estimate of the unknown tangential velocities of these galaxies. Dashed curves show the escape velocity from a 10¹² M_☉ point mass (a reasonable approximation to the escape velocity curve of the MW for the purposes of this discussion). The vertical dashed line indicates the approximate cosmological virial radius of the MW, R_vir ∼ 300 kpc (Klypin et al. 2002; these authors define R_vir as the radius within which the mean density of the model dark matter halo is a factor Δ ≃ 340 larger than the critical density of the universe).

Zoom In Zoom Out Reset image size

With the obvious caveat regarding the generally unknown tangential components of the velocities of the galaxies, Figure 2 can be used to define membership of the MW subgroup. In particular, all galaxies within 300 kpc of the MW are likely bound satellites. This is with the exception of Leo I, which may be unbound unless its tangential components are significantly overestimated (also see Mateo et al. 2008; Sohn et al. 2007; Zaritsky et al. 1989, and references therein). Leo T, NGC 6822, and Phoenix may all be bound in this simple model, but their large distance from the MW makes their membership of the MW subgroup ambiguous. I note that no potential satellite of the MW or M31 (Section 3.2) is clearly unbound as a result of the magnitude of its radial velocity alone; all the potential satellites with large radial velocities relative to their host may be bound if their tangential velocities are sufficiently small. For the purpose of the classification in Column 3 of Table 1, and in the forthcoming discussion, I only include as definite MW satellites those objects that are bound by the dashed curves in Figure 2 and which have D_G ⩽ R_vir. It is interesting to note that the radial distribution of satellites is such that they are found at all radii out to ∼280 kpc from the MW, at which point there is a notable gap in the distribution until the next set of objects at >400 kpc. A similar gap is present at a similar radius for satellites around M31 (Section 3.2; but see the recent discovery of Andromeda XXVIII by Slater et al. 2011). This tentatively suggests that this gap could be used to observationally define the limiting radius of the sub-groups (or equivalently, the extent of the two host galaxies) and appears consistent with previous work and other galaxy groups (e.g., Cen A, Sculptor, IC 342: Karachentsev et al. 2002a, 2003a, 2003b; see also Grebel et al. 2003).

Pisces II is the most recently discovered satellite to the MW (Belokurov et al. 2010) and does not appear in Figure 2 since it currently lacks a velocity measurement. At the time of writing, 15 new satellites have been discovered since 2005, all of them using the Sloan Digital Sky Survey (SDSS) and hence confined to its footprint. Irwin (1994) originally argued that the MW sub-group was essentially complete at Galactic latitudes away from the disk for objects of comparable luminosity as (or brighter than) Sextans. The fact that only one of the new satellites (Canes Venatici) is comparable in luminosity to any of the previously known MW satellites backs up this claim. Indeed, Koposov et al. (2008) demonstrate that most of the newly discovered galaxies are at the limits of detection of SDSS, implying that—even within the SDSS footprint—a large number of galaxies still await discovery. This was explored further by Tollerud et al. (2008), who conclude that there may by many hundreds of unseen faint satellites.

It is worth emphasizing the problems caused by the Galactic disk in searching for satellites. Figure 3 shows the distribution of MW satellites with (absolute) latitude. The histogram shows the observed distribution of all 27 satellites, whereas the dashed line indicates a uniform distribution, normalized to match observations at |b| > 30°. There are three satellites at |b| < 30° (one of which is Canis Major, on which there is still considerable debate as to whether it is a dwarf galaxy or an unrelated disk substructure; e.g., Martin et al. 2004a, 2004b; Momany et al. 2004; Martínez-Delgado et al. 2005; Moitinho et al. 2006; Butler et al. 2007; López-Corredoira et al. 2007), a number that is very low considering this accounts for half of the sky. Obscuration by the Galactic disk, however, does not become a serious problem until |b| ≲ 20°. While some of the area probed by SDSS reaches to very low Galactic latitudes, the majority of the Legacy area is located at b > 30°, likely explaining the dearth of satellites at intermediate latitudes.

Zoom In Zoom Out Reset image size

Two photometric surveys currently underway—the Pan-STARRS (Kaiser et al. 2002) PS1 survey in the north and SkyMapper (Keller et al. 2007) in the south—together will map the entire sky to equivalent or deeper magnitude limits to SDSS and consequently should reveal the presence of many more satellites. These should remove the spatial bias introduced by SDSS. In addition, other wide-field (but not all-hemisphere) surveys such as that made possible with the soon-to-be-commissioned Subaru/HyperSuprimeCam will allow for even fainter satellites to be detected (if they exist). In the more distant future, the Large Synoptic Survey Telescope will re-examine the southern hemisphere sky to even greater depth. Judging by recent discoveries, there is every reason to expect that large numbers of very faint satellites will be discovered by all these projects. However, objects behind the Galactic plane will likely remain hidden, and it is unlikely that (m)any bright (M_V < −8) Galactic dwarf galaxies at higher Galactic latitudes remain to be discovered.

3.2. Membership of the M31 Sub-group

Figure 4 shows M31-centric velocity versus distance for all galaxies within 600 kpc of Andromeda for which these measurements exist. Again, the M31-centric velocity listed in Table 2 has been multiplied by a factor of $\sqrt{3}$ to account for the unknown tangential velocities of these galaxies. Following Figure 2, dashed curves show the escape velocity from a 10¹² M_☉ point mass, and the vertical dashed line indicates the approximate cosmological virial radius of M31, R_vir ∼ 300 kpc (e.g., Klypin et al. 2002).

Zoom In Zoom Out Reset image size

Whereas only Leo I around the MW stands out as a possible unbound satellite from its radial velocity (Mateo et al. 2008), M31 has at least two close-in dwarf galaxies that are likely unbound, Andromeda XII (Chapman et al. 2007) and XIV (Majewski et al. 2007). The Pegasus dwarf irregular galaxy (DIG) and IC 1613 are both extremely distant from Andromeda but could arguably be dynamically associated. In this respect they are similar to NGC 6822, LeoT, and Phoenix around the MW. It is often overlooked that IC 10 and LGS 3 are clear M31 satellites (as close to M31 as the Andromeda VI and XVI dSphs). However, these two galaxies differ from the majority of satellites insofar as they both have non-negligible gas fractions and younger stellar populations. LGS 3 is generally classified as being morphologically akin to LeoT (a "transition-type" dwarf; see Section 5 for discussion), whereas IC 10 is generally classified as a low-mass dwarf irregular. Indeed, as a whole, the M31 satellite system contrasts with the MW satellites in the variety of morphological types present. Around the MW, only the LMC and SMC are not classified as dSphs. However, in addition to the dSph satellites, M31 also has a transition dwarf (LGS 3), a dwarf irregular (IC 10, although it is much lower luminosity than either of the Magellanic Clouds), three dwarf elliptical galaxies (NGC 205, 147, and 185), a compact elliptical (M32), and a low-mass spiral galaxy (M33).

In the past seven years we have seen a rapid growth in the number of known M31 satellites through focused wide-field studies of the environment around Andromeda. All the new dwarfs have been discovered as overdensities of red giant branch (RGB) stars. This, therefore, imposes a basic magnitude limit on the galaxies that can be found, since it requires galaxies to have enough stars that a reasonable number are passing through the RGB phase. An exact study of the selection effects for dwarf galaxy searches around M31 has yet to be published, but the practical limit of these searches appears to be around M_V ∼ −6. The most extensive survey of M31's environs, by the Pan-Andromeda Archaeological Survey (PAndAS; McConnachie et al. 2009), provides complete spatial coverage out to a maximum projected radius from M31 of 150 kpc. As shown in McConnachie et al. (2009) and Richardson et al. (2011), the projected radial distribution of satellites shows no sign of declining within this radius. It is highly probable, therefore, that a large number of satellites remain to be discovered at large radius (recently verified by the discoveries of Andromeda XXVIII (Slater et al. 2011) and Andromeda XXIX (Bell et al. 2011) in the SDSS DR8). At small radius (≲ 60 kpc), there is also a deficit of galaxies due to the dominance of M31 stellar populations in this region, making it difficult to identify any low-luminosity satellites that may be superimposed in front of (or behind) the main body of M31.

Within both the MW and the M31 sub-group there are at least three potential dynamical associations of satellites ("sub-sub-groups"). Around the MW, Leo IV and V are in close proximity to each other (positions and velocities) and may well constitute a binary system (Belokurov et al. 2008; de Jong et al. 2010). Within the M31 sub-group, Andromeda XXII is only about 40 kpc in projection from M33, whereas it is over 200 kpc in projection from M31 (Martin et al. 2009). Thus, Andromeda XXII may actually be the only known satellite of M33 (the bright dSph Andromeda II is also considerably closer to M33 than M31, but in a region where the gravitational potential of M31 clearly dominates). A second possible pairing is found to the north of M31 and includes NGC 185 and 147. Van den Bergh (1998) postulated that NGC 185 and 147 were a binary system; they are separated by only 1° in projection and their distances imply a three-dimensional separation of ∼60 kpc. Their velocities differ by only ∼10 km s⁻¹, and such a small separation in phase-space seems unlikely (but not impossible) if these galaxies are not, or never have been, a binary system.

3.3. Membership of the Local Group, and Identification of Nearby Neighbors

Figure 5 shows Local-Group-centric velocity versus distance from the barycenter of the Local Group for all galaxies in the sample for which these data exist. The Local-Group-centric velocity listed in Table 2 has been multiplied by a factor of $\sqrt{3}$ to account for the unknown tangential components. Dashed curves indicate the escape velocity from a point mass of 2 × 10¹² M_☉. Dotted curves indicate the escape velocity from a point mass of 5 × 10¹² M_☉, the approximate total dynamical mass of the Local Group as implied from the timing argument (e.g., Lynden-Bell 1981). Many of the points at D_LG ≲ 500 kpc represent galaxies that are satellites of either the MW or M31. These systems are dynamically dominated by the gravitational potential of these massive hosts rather than the Local Group as a whole.

Zoom In Zoom Out Reset image size

There is a clear separation in Figure 5 between outer, apparently bound, Local Group members (the outermost of which is UGC 4879) and the next nearest galaxies (the four members of the NGC 3109 grouping; see van den Bergh 1999). The rise in velocity with distance for all the nearby neighbors is due to the Hubble expansion, and the reader is referred to Karachentsev et al. (2009) for a beautiful analysis of the Hubble flow around the Local Group.

The six outermost Local Group members cluster around v_LG = 0, and their mean velocity is consistent with zero (〈v〉 = 4 ± 16 km s⁻¹, where the uncertainty is the random error in the mean). As such, their mean distance can be used as an estimate of the zero-velocity surface of the Local Group and is R_LG = 1060 ± 70 kpc (where the uncertainty is the random error in the mean; cf. Courteau & van den Bergh 1999; Karachentsev et al. 2002b, 2009; and others). It is also worth pointing out that the velocity dispersion of the Local Group can be estimated using those Local Group galaxies that are far from either the MW or M31 and that have published velocities. The six Local Group galaxies identified in Figure 5, with the addition of IC 1613 and Cetus, are all more than 500 kpc from either of the two large galaxies and yield σ_LG = 49 ± 36 km s⁻¹ (where the uncertainty is the random error in the dispersion). Note that there is a systematic error component in all these estimates caused by our choice for the location of the barycenter of the Local Group, although it has a generally weak effect on any values given here. The fact that the measured velocity dispersion for the entire Local Group is so low that it is of order the velocity dispersion of the gas in the SMC is peculiar and well noted (e.g., Sandage 1986, and references therein). It may result from the necessity to measure the dispersion from galaxies that are mostly at or near turnaround.

Not including satellites of the MW and M31, it is highly likely that the census of Local Group galaxies remains severely incomplete. Unlike recently discovered MW satellites, isolated Local Group galaxies are not nearby. Unlike the M31 sub-group, isolated Local Group galaxies do not cluster in a specific area of sky amenable to dedicated surveys. The middle panel of Figure 1 shows that 8 of the 13 (quasi-)isolated Local Group galaxies are all in one quadrant of the sky (the same one that contains M31). However, it is also not clear what the expected spatial distribution of non-satellite Local Group galaxies should be, and it is likely naive to expect an isotropic distribution. Heroic searches for new Local Group galaxies and nearby neighbors have been made by groups led by I. Karachentsev, V. Karachentseva (e.g., Karachentseva & Karachentsev 1998; Karachentseva et al. 1999; Karachentsev et al. 2000 and many others), and A. Whiting (Whiting et al. 1997, 1999) through the visual inspection of literally every plate of the second Palomar Observatory Sky Survey and the ESO/Science Research Council survey, and these remain the most comprehensive all-sky searches to date. Whiting et al. (2007) attempt to quantify their visual search of these plates; they estimate that their search is ∼80% complete in the region away from interference from the MW disk (more than 70% of the sky) to a surface brightness level approaching 26 mag arcsec⁻². This translates to (at most) 1 or 2 currently uncataloged Local Group members down to this limit, away from the disk of the Galaxy. Current and future large-scale digital surveys of the sky—including blind H i surveys as well as optical studies—are potentially rich hunting grounds to try to improve the census of nearby galaxies, be they members of the Local Group or the nearby neighbors that help define its environment.

Table 3 presents global photometric and structural parameters for each galaxy in the sample.

Table 3. Luminosity and Structural Parameters

(1)	(2)		(3)		(4)		(5)		(6)		(7)		(8)		(9)	(10)	(11)
Galaxy	V		rh		μV		θ				MV		rh		μeff	Referencesa	Comments
			(')		(M/□'')		(°)						(pc)		(M/□'')
The Galaxy
Canis Major	−0.1	0.8	...	...	24.0b	0.6	123:c		0.8:		−14.4d	0.8	...	...	...	(1) (2)
Sagittarius dSph	3.6	0.3	342.00	12.00	25.2	0.3	102	2	0.64	0.02	−13.5	0.3	2587	219	26.0	(3) (4)
Segue	15.3	0.8	4.40	+1.2−0.6	27.6	1.0	85	8	0.48	0.13	−1.5	0.8	29	+8−5	28.7	(5)
Ursa Major II	13.3	0.5	16.00	1.00	27.9	0.6	98	4	0.63	0.05	−4.2	0.6	149	21	29.1	(5)
Bootes II	15.4	0.9	4.20	1.40	28.1	1.6	145	55	0.21	0.21	−2.7	0.9	51	17	29.1	(5)
Segue II	15.2	0.3	3.40	0.20	27.4	0.4	182	17	0.15	0.10	−2.5	0.3	35	3	28.6	(6)
Willman 1	15.2	0.7	2.30	0.40	26.1	0.9	77	5	0.47	0.08	−2.7	0.8	25	6	27.2	(5)
Coma Berenices	14.1	0.5	6.00	0.60	27.3	0.7	115	10	0.38	0.14	−4.1	0.5	77	10	28.4	(5)
Bootes III	12.6	0.5	...	...	31.3e	0.3	90:		0.50:		−5.8d	0.5	...	...	...	(7)
LMC	0.4	0.1									−18.1	0.1				(8)
SMC	2.2	0.2									−16.8	0.2				(8)
Bootes	12.8	0.2	12.60	1.00	27.5	0.3	14	6	0.39	0.06	−6.3	0.2	242	21	28.7	(5)
Draco	10.6	0.2	10.00	0.30	25.0	0.2	89	2	0.31	0.02	−8.8	0.3	221	19	26.1	(5)
Ursa Minor	10.6	0.5	8.20	1.20	26.0	0.5	53	5	0.56	0.05	−8.8	0.5	181	27	25.2	(9)
Sculptor	8.6	0.5	11.30	1.60	23.5	0.5	99	1	0.32	0.03	−11.1	0.5	283	45	24.3	(9)
Sextans	10.4	0.5	27.80	1.20	27.1	0.5	56	5	0.35	0.05	−9.3	0.5	695	44	28.0	(9)
Ursa Major	14.4	0.3	11.30	1.70	27.7	0.5	71	3	0.80	0.04	−5.5	0.3	319	50	28.8	(5)
Carina	11.0	0.5	8.20	1.20	25.5	0.5	65	5	0.33	0.05	−9.1	0.5	250	39	26.0	(9)
Hercules	14.0	0.3	8.60	+1.8−1.1	27.2	0.6	102	4	0.68	0.08	−6.6	0.4	330	+75−52	28.3	(5)
Fornax	7.4	0.3	16.60	1.20	23.3	0.3	41	1	0.30	0.01	−13.4	0.3	710	77	24.0	(9)
Leo IV	15.1	0.4	4.60	0.80	27.5	0.7	121	9	0.49	0.11	−5.8	0.4	206	37	28.6	(10)
Canes Venatici II	16.1	0.5	1.60	0.30	26.1	0.7	177	9	0.52	0.11	−4.9	0.5	74	14	27.2	(5)
Leo V	16.0	0.4	2.60	0.60	27.1	0.8	96	13	0.50	0.15	−5.2	0.4	135	32	28.2	(10)
Pisces II	16.3	0.5	1.10	0.10	25.7	0.6	77	12	0.40	0.10	−5.0:		58	99	26.8	(11)
Canes Venatici	13.1	0.2	8.90	0.40	27.1	0.2	70	4	0.39	0.03	−8.6	0.2	564	36	28.2	(5)
Leo II	12.0	0.3	2.60	0.60	24.2	0.3	12	10	0.13	0.05	−9.8	0.3	176	42	24.8	(9)
Leo I	10.0	0.3	3.40	0.30	22.6	0.3	79	3	0.21	0.03	−12.0	0.3	251	27	23.3	(9)
Andromeda
M32	8.1	0.1	0.47	0.05	11.1:f		159	2	0.25	0.02	−16.4	0.2	110	16	17.0	(8) (12)
Andromeda IX	16.3	1.1	2.50	0.10	28.0	1.2	...	...	...	...	−8.1	1.1	557	29	29.2	(13) (14)
NGC 205	8.1	0.1	2.46	0.10	15.4:g		28	5	0.43	0.10	−16.5	0.1	590	31	20.3	(8) (12)
Andromeda XVII	15.8	0.4	1.24	0.08	25.7	0.5	122	7	0.27	0.06	−8.7	0.4	286	23	26.8	(15)
Andromeda I	12.7	0.1	3.10	0.30	24.7	0.2	22	15	0.22	0.04	−11.7	0.1	672	69	25.8	(16)
Andromeda XXVIIh	16.7	0.5	1.80	0.30	27.6	0.5	150	10	0.40	0.20	−7.9	0.5	434	76	28.3	(17)
Andromeda III	14.4	0.3	2.20	0.20	24.8	0.2	136	3	0.52	0.02	−10.0	0.3	479	46	26.2	(16)
Andromeda XXV	14.8	0.5	3.00	0.20	27.1	0.5	170	10	0.25	0.05	−9.7	0.5	709	61	27.8	(17)
Andromeda XXVI	17.3	0.5	1.00	0.10	27.4	0.5	145	10	0.25	0.05	−7.1	0.5	222	25	27.9	(17)
Andromeda XI	17.5	1.2	0.71	0.03	26.5	1.3	...	...	...	...	−6.9	1.3	157	+16−37	27.6	(13) (14)
Andromeda V	15.3	0.2	1.40	0.20	25.3	0.2	32	10	0.18	0.05	−9.1	0.2	315	46	26.7	(16)
Andromeda X	16.6	1.0	1.30	0.10	26.3	1.1	46	5	0.44	0.06	−7.6	1.0	265	33	27.4	(15)
Andromeda XXIII	14.2	0.5	4.60	0.20	28.0	0.5	138	5	0.40	0.05	−10.2	0.5	1029	76	27.8	(17)
Andromeda XX	18.2	0.8	0.53	+0.14−0.04	26.2	0.8	80	20	0.30	0.15	−6.3	+1.1−0.8	124	+53−17	27.3	(18)
Andromeda XII	18.3	1.2	1.20	0.20	28.5	1.3	...	...	...	...	−6.4	1.2	304	66	29.6	(13) (14)
NGC 147	9.5	0.1	3.17:		21.2:		25	3	0.41	0.02	−14.6	0.1	623:		22.3	(8)	r25 = 6.6 ± 0.3 arcmins
Andromeda XXI	14.8	0.6	3.50	0.30	27.0	0.4	110	15	0.20	0.07	−9.9	0.6	875	91	28.2	(19)
Andromeda XIV	15.9	0.5	1.70	0.80	27.2	0.6	...	...	0.31	0.09	−8.4	0.6	363	180	27.5	(20)
Andromeda XV	14.6	0.3	1.21	0.05	24.8	0.4	0:		0.00:		−9.4	0.4	222	22	25.9	(21)
Andromeda XIII	18.1	1.2	0.78	0.08	27.3	1.3	...	...	...	...	−6.7	1.3	207	+23−44	28.4	(13) (14)
Andromeda II	11.7	0.2	6.20	0.20	24.5	0.2	34	6	0.20	0.08	−12.4	0.2	1176	50	26.3	(16)
NGC 185	9.2	0.1	2.55:		20.8:		35	3	0.15	0.01	−14.8	0.1	458:		21.9	(8)	r25 = 5.8 ± 0.3 arcmins
Andromeda XXIX	16.0	0.3	1.70	0.20	27.6	0.4	51	8	0.35	0.06	−8.3	0.4	361	56	27.6	(22)
Andromeda XIX	15.6	0.6	6.20	0.10	29.3	0.7	37	8	0.17	0.02	−9.2	0.6	1683	105	30.2	(18)
Triangulum	5.7	0.1					23		0.41		−18.8	0.1				(8)
Andromeda XXIV	16.3	0.5	2.10	0.10	27.8	0.5	5	10	0.25	0.05	−7.6	0.5	367	27	28.5	(17)
Andromeda VII	11.8	0.3	3.50	0.10	23.2	0.2	94	8	0.13	0.04	−12.6	0.3	776	42	25.3	(16)
Andromeda XXII	18.0	0.8	0.94	0.10	26.7	0.6	114	15	0.56	0.11	−6.5	9.9	217	99	27.9	(19)
IC 10	9.5	0.2	2.65:		24.6	0.2	...	...	0.19	0.02	−15.0	0.2	612:		22.3	(8) (23)	r25 = 3.2 ± 0.1 arcmins
LGS 3	14.3	0.1	2.10	0.20	24.8	0.1	0:		0.20:		−10.1	0.1	470	47	26.6	(24)
Andromeda VI	13.2	0.2	2.30	0.20	24.1	0.2	163	3	0.41	0.03	−11.3	0.2	524	49	25.3	(16)
Andromeda XVI	14.4	0.3	0.89	0.05	23.9	0.4	0:		0.00:		−9.2	0.4	136	15	25.0	(21)
Andromeda XXVIII	15.6	+0.4−0.9	1.11	0.21	26.3	+0.4−0.9	39	16	0.34	0.13	−8.5	+0.4−1.0	213	+64−45	26.3	(25)
IC 1613	9.2	0.1	6.81:		22.8:		50	2	0.11	0.03	−15.2	0.2	1496:		24.1	(8)	r25 = 8.1 ± 0.2 arcmins
Phoenix	13.2	0.4	3.76:		25.8:		5i	1	0.40j	0.10	−9.9	0.4	454:		26.4	(26) (27)	rt = 15.8+4.3−2.8 arcmins; assumed c = 0.7
NGC 6822	8.1	0.2	2.65:		19.7:		330	10	0.24	0.05	−15.2	0.2	354:		20.8	(8) (28)	r25 = 7.7 ± 0.2 arcmins
Cetus	13.2	0.2	3.20	0.10	25.0	0.2	63	3	0.33	0.06	−11.2	0.2	703	31	26.2	(16)
Pegasus dIrr	12.6	0.2	2.10:		22.8:		120	2	0.46	0.02	−12.2	0.2	562:		24.4	(8)
Leo T	15.1k	0.5	0.99	0.06	24.8	0.6	0:		0.00:		−8.0	0.5	120	9	26.0	(29)
WLM	10.6	0.1	7.78:		23.7:		4	2	0.65	0.01	−14.2	0.1	2111:		24.8	(8)
Leo A	12.4	0.2	2.15:		22.8:		114	5	0.40	0.03	−12.1	0.2	499:		24.4	(8) (30)	r25 = 2.6 ± 0.1 arcmins
Andromeda XVIII	16.0	9.9	0.92	0.06	25.6:		...	...	...	...	−9.7:		363	32	26.7	(18)	Lower limits on V and μV
Aquarius	14.5	0.1	1.47	0.04	23.6	0.2	99	1	0.50	0.10	−10.6	0.1	458	21	25.5	(31)
Tucana	15.2	0.2	1.10	0.20	25.0	0.1	97	2	0.48	0.03	−9.5	0.2	284	54	25.6	(32)
Sagittarius dIrr	13.6	0.2	0.91	0.05	23.9	0.2	90:		0.50:		−11.5	0.3	282	28	23.5	(33)	rh from integral under central King profile
UGC 4879	13.2	0.2	0.41	0.04	21.2	0.2	84	11	0.44	0.04	−12.5	0.2	162	16	21.5	(34)
NGC 3109	10.7	0.1	4.30	0.10	22.6	0.1	92	1	0.82	0.01	−14.9	0.1	1626	71	22.9	(35)
Sextans B	11.3	0.2	1.06	0.10	21.9	0.3	110	2	0.31	0.03	−14.5	0.2	440	42	21.9	(8) (36)
Antlia	15.2	0.2	1.20	0.12	23.9	0.2	135	5	0.40	0.04	−10.4	0.2	471	52	25.9	(37) (36)
Sextans A	11.5	0.1	2.47:		22.8:		0	1	0.17	0.02	−14.3	0.1	1029:		24.1	(8) (38)	r25 = 2.9 ± 0.1 arcmins
HIZSS 3(A)	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
HIZSS 3B	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
KKR 25	15.9	0.2	0.40:		24.0	0.2	...	...	0.41	0.02	−10.5	0.2	222:		24.2	(39)	r26.5 = 0.55 arcmins
ESO 410- G 005	14.9	0.3	0.50	0.05	22.2	0.2	57:		0.37:		−11.5	0.3	280:		23.8	(40) (36)
NGC 55	7.9	0.1	5.16:		19.3:		108	2	0.83	0.01	−18.5	0.2	2900:		20.4	(8)	r25 = 16.2 ± 0.4 arcmins
ESO 294- G 010	15.3	0.3	0.42	0.04	22.3	0.2	8:		0.51:		−11.2	0.3	248	24	23.5	(40) (36)
NGC 300	8.1	0.1	5.00:		21.0:		111	2	0.29	0.01	−18.5	0.1	3025:		22.1	(8)	r25 = 10.1 ± 0.3 arcmins
IC 5152	10.9:l		0.97:		20.1:		100	2	0.38	0.02	−15.6:		550:		21.2	(8)	r25 = 2.6 ± 0.1 arcmins
KKH 98	15.2	0.3	0.64	0.06	22.8	0.2	−5	1	0.41	0.01	−11.8	0.3	470	48	24.5	(36) (38)
UKS 2323-326	13.5	0.2	0.90	0.10	22.9	0.2	−60	4	0.10	0.01	−13.2	0.2	578	69	24.0	(41) (38)
KKR 3	17.2	0.3	0.36	0.04	23.8	0.2	0	1	0.05	0.01	−9.5	0.3	229	28	25.8	(36) (38)
GR 8	14.5	0.2	0.32	0.04	22.6	0.2	61	2	0.20	0.05	−12.2	0.2	203	28	22.7	(42) (43)
UGC 9128	14.4	0.3	0.64	0.07	22.6	0.2	46	2	0.40	0.05	−12.4	0.3	427	47	23.8	(43) (36)
UGC 8508	13.7	0.1	0.42	0.04	21.5	0.2	−60	2	0.45	0.05	−13.4	0.1	315	30	22.1	(43) (36)
IC 3104	12.8:l		2.01:		23.3:		45	2	0.52	0.02	−14.0:		1327:		24.4	(8)	r25 = 1.9 ± 0.1 arcmins
DDO 125	12.7	0.3	1.04	0.10	22.1	0.2	−68	4	0.41	0.01	−14.4	0.3	781	77	23.1	(36) (38)
UGCA 86	14.2:l		0.94:		22.8:		25	1	0.32	0.03	−13.2:		811:		24.5	(44) (38)	r25 = 2.3 ± 0.2 arcmins
DDO 99	13.9	0.1	0.90	0.09	22.9	0.2	70	4	0.29	0.01	−13.2	0.2	679	81	24.2	(42) (38)
IC 4662	11.1	0.3	0.48	0.05	18.7	0.2	−69	4	0.27	0.01	−15.8	0.3	341	44	20.1	(36) (38)
DDO 190	12.8	0.1	0.64	0.06	21.4	0.2	82	5	0.10	0.02	−14.4	0.1	520	49	22.6	(45) (36)
KKH 86	17.1	0.3	0.28	0.03	23.2	0.2	−3	1	0.39	0.01	−10.0	0.3	210	27	24.7	(36) (38)
NGC 4163	13.2	0.3	0.45	0.05	21.1	0.2	11	2	0.30	0.05	−14.1	0.3	374	42	22.0	(43) (36)
DDO 113	16.4	0.3	0.70	0.07	24.0	0.2	0:		0.00	0.09	−11.0	0.3	601	62	26.5	(8) (36)

^aReferences: (1) Bellazzini et al. 2006; (2) Butler et al. 2007; (3) Mateo et al. 1998; (4) Majewski et al. 2003; (5) Martin et al. 2008b; (6) Belokurov et al. 2009; (7) Correnti et al. 2009; (8) de Vaucouleurs et al. 1991; (9) Irwin & Hatzidimitriou 1995; (10) de Jong et al. 2010; (11) Belokurov et al. 2010; (12) Choi et al. 2002; (13) Collins et al. 2010; (14) Collins 2011; (15) Brasseur et al. 2011b; (16) McConnachie & Irwin 2006; (17) Richardson et al. 2011; (18) McConnachie et al. 2008; (19) Martin et al. 2009; (20) Majewski et al. 2007; (21) Ibata et al. 2007; (22) Bell et al. 2011; (23) Sanna et al. 2010; (24) Lee 1995; (25) Slater et al. 2011; (26) van de Rydt et al. 1991; (27) Martínez-Delgado et al. 1999; (28) Dale et al. 2007; (29) de Jong et al. 2008; (30) Vansevičius et al. 2004; (31) McConnachie et al. 2006; (32) Saviane et al. 1996; (33) Lee & Kim 2000; (34) Bellazzini et al. 2011; (35) Hidalgo et al. 2008; (36) Sharina et al. 2008; (37) Aparicio et al. 1997; (38) Fingerhut et al. 2010; (39) Karachentsev et al. 2001b; (40) Loveday 1996; (41) Lee & Byun 1999; (42) Makarova 1999; (43) Vaduvescu et al. 2005; (44) Karachentsev et al. 1997; (45) Aparicio & Tikhonov 2000. ^bMeasured within central 2 × 2 deg. ^cBroadly aligned with the Galactic Plane. ^dDistance-independent absolute magnitude estimate. ^eMeasured within projected radius of 0.8 deg. ^fB magnitude, corresponding to "standard fit" in Table 1 of Choi et al. (2002). ^gB magnitude; see Table 1 of Geha et al. (2006b). ^hMeasurement of structural parameters complicated by presence of surrounding stream. ⁱCorresponds to "outer component"; θ_inner = 95°. ^jCorresponds to "outer component"; _inner = 0.4. ^kConverted from g, r. ^lB magnitude.

Download table as:  ASCIITypeset images: 1 2 3

Column 1. Galaxy name.

Column 2. Apparent magnitude in V (Vega magnitudes). In general, these have been corrected for foreground extinction (but not internal extinction), although in some cases the original papers are unclear as to whether extinction corrections were applied to the apparent magnitudes. For many of the fainter galaxies that have been resolved into stars, the magnitude may not be based on integrated light (often undetectable) and is instead estimated by measuring the luminosity of the brighter stars and by making assumptions regarding the luminosity function of stars below the detection limit.

Column 3. Radius (arcmin) containing half the light of the galaxy, measured on the semimajor axis. A vast array of different scale radii are used in the literature, and in cases where the original papers use other scale radii (e.g., core, exponential, half-brightness, R₂₅, etc.), I have derived half-light radii by integrating under the appropriate profile, normalized to the apparent magnitude (and with the appropriate ellipticity if measured, otherwise assumed to be circular). In cases where the original papers did not fit a profile, I have assumed an exponential profile. The latter estimates are particularly uncertain, and in these cases I have also given the original scale radius in Column 11.

Column 4. Central surface brightness (mag arcsec⁻²). Where possible, I cite directly measured values, although in many cases (including in the original papers) this parameter is derived by normalizing the measured profile to the apparent magnitude.

Column 5. Position angle of major axis, measured in degrees east from north. For systems that show a change in this quantity as a function of radius, a mean value has been estimated.

Column 6. Ellipticity = 1 − b/a, where b is the semiminor axis and a is the semimajor axis. For systems that show a change in this quantity as a function of radius, a mean value has been estimated.

Column 7. Absolute visual magnitude, derived from Column 2 by subtraction of the distance modulus given in Table 2.

Column 8. Half-light radius in parsecs, derived from Column 3 assuming the distance modulus given in Table 2.

Column 9. The mean surface brightness within the isophote defined by the half-light radius, derived using values from Columns 2, 3, and 6. In cases where no ellipticity is measured, I have assumed circular isophotes.

Column 10. References.

Column 11. Comments.

For each galaxy I have tried to minimize the number of different papers that I cite to provide complete information, with the result that Table 3 references 45 papers from the last 21 years. Concern must be given to the systematic uncertainties that exist between measurements: they are necessarily complex, they are essentially unquantifiable, and they probably present the primary limitation for examination of the global properties of this population.

Despite the usefulness of parameters such as total luminosity and half-light radius, it should be strongly stressed that these numbers are wholly inadequate in reflecting the known complexity of galaxy structures. It is standard practice to decompose bright galaxies into components such as nucleus and envelope, or a bulge, a disk (several disks?), and a halo. More recently, however, a vast number of independent studies of galaxies included in Table 3 have shown that a full description of their structure requires decomposition of their surface brightness profile into multiple components (e.g., Irwin & Hatzidimitriou 1995; Martínez-Delgado et al. 1999; Lee et al. 1999; Lee & Byun 1999; Lee & Kim 2000; Vansevičius et al. 2004; McConnachie et al. 2007; Hidalgo et al. 2008, among many others). The addition of spectroscopic data (see Section 5) adds new complexity to the discussion as well, since dynamically distinct (e.g., Kleyna et al. 2003, 2004) and chemo-dynamically distinct (e.g., Tolstoy et al. 2004; Battaglia et al. 2006, 2008, 2011) populations of stars have been shown to reside within individual dwarf galaxies that do not necessarily always reveal themselves as features in the global surface brightness profiles. Thus, not only are integrated photometric properties such as r_h and M_V inadequate to describe the complex structures exhibited by many galaxies, but so too are single velocity dispersion or rotation measurements (Section 5; true for H i as well as stars; e.g., Lo et al. 1993; Young & Lo 1996, 1997a, 1997b; Young et al. 2003).

Figures 6 and 7 show basic scaling relations using the galaxy parameters given in Table 3, first studied in detail in Kormendy (1985). In Figure 6, I plot absolute magnitude against half-light radius. Here, I have converted the half-light radius given in Table 3 to the geometric mean half-light radius of the major and minor axes ($r = \sqrt{r_ar_b}$, where r_a and r_b are radii measured on the major and minor axes, respectively) to account for the presence of highly elliptical systems. The top panel includes as small dots the location of the Galactic globular clusters, from the data compiled by Harris (1996), and the dashed line shows the direction defined by points of constant surface brightness (averaged within the half-light radius). I have labeled obvious outlying points, and I have also labeled some of the least luminous (candidate) dwarf galaxies. The lower panel shows an expansion of the region −17 ⩽ M_V ⩽ −7.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The observed relationship between absolute magnitude and half-light radius shown in the top panel of Figure 6 spans a factor of one million in luminosity. The galaxies seem well separated from the (Galactic) globular clusters in this two-dimensional projection, although the limited number of low-luminosity galaxies (and the fact that all of them are MW satellites) complicates interpretation of this region of the diagram. When M_V and r_h are considered separately, the one-dimensional distributions do overlap, such that the smallest, least luminous (candidate) dwarf galaxies are as faint as any known globular clusters and smaller than the most extended clusters. The distinction between these stellar systems in this parameter space becomes less clear when other types of compact stellar systems, such as ultracompact dwarfs and the nuclei of early-type galaxies, are included (e.g., see Figures 5 and 8 of Brodie et al. 2011).

Three galaxies are clear outliers in the top panel of Figure 6: Andromeda XIX is extremely extended for its luminosity (see also Figure 7), whereas the compact elliptical M32 is a known outlier in this projection due to its extreme concentration (see also Kormendy & Bender 2012, and references therein). Interestingly, the recently identified Local Group galaxy UGC 4879 (Kopylov et al. 2008) tends to the same side of the relation as M32, although not as extreme. Bellazzini et al. (2011) note that UGC 4879 has an unusual structure, insofar as there appear to be two flattened wings emanating from the central spheroid of stars, perhaps indicating the presence of a disk. More work is needed to determine the significance of the structure of UGC 4879 and its position in this diagram.

The lower panel of Figure 6 enlarges the region −17 ⩽ M_V ⩽ −7 to demonstrate the very large scatter that exists in this regime (see below for discussion of the fainter systems). McConnachie & Irwin (2006) originally pointed out that, at a given luminosity, the M31 dwarf satellites were generally larger than the MW population. With a large number of new discoveries since this time, Brasseur et al. (2011a) demonstrate that the mean relations of the M31 and MW populations are statistically consistent with each other³. The limited number of points contributing to the relations for these different sub-groups makes interpretation difficult; for example, 4 out of 10 MW satellites have r_h > 250 pc, whereas 20 out of 26 M31 satellites are this large. It may be that local environment contributes to determining the sizes of dwarf galaxies, but perhaps the most important conclusion that can be drawn from the lower panel of Figure 6 is the lack of a strong relation between M_V–r_h in this regime; at any given luminosity, galaxy sizes vary by nearly an order of magnitude, and there is only a weak dependence on luminosity.

At even fainter magnitudes, the line of constant surface brightness (averaged within the half-light radius) included in the top panel of Figure 6 runs approximately parallel to the relation defined by the galaxy locus for M_V ≳ −8 (and this also has a different slope than the weak relation at brighter magnitudes). This indicates that the smallest galaxies so far discovered all have similar surface brightnesses and further suggests that selection effects may be driving the form of the M_V − r_h relation at faint magnitudes. To explore this further, I plot surface brightness against absolute magnitude in Figure 7. The top panel shows the central surface brightness, whereas the bottom panel uses the average surface brightness within the half-light radius. The former is in principle more susceptible to localized features (e.g., nuclei, individual bright stars, etc.) than the latter. I have labeled points that are obvious outliers in both panels (note that Bootes III is not present in the lower panel of Figure 7 due to the lack of a measured value for r_h). These galaxies are all known or suspected to be undergoing tidal disruption.

There is a clear correlation between absolute magnitude and surface brightness in both panels of Figure 7 for galaxies more luminous than M_V ∼ −9. Of considerable interest, however, is the change in the observed relationship between surface brightness and absolute magnitude within the dwarf regime. In particular, dwarf galaxies fainter than M_V ≃ −8.5 do not continue to decrease in surface brightness with decreasing luminosity. Instead, the surface brightness levels off at faint magnitudes, as implied in Figure 6. The central surface brightness distribution for the 26 galaxies fainter than M_V = −8.5 in Figure 7 can be characterized by a median surface brightness of 27.35 mag arcsec⁻² and an inter-quartile range of 1.2 mag. Thus, the central surface brightness for galaxies fainter than M_V = −8.5 appears remarkably constant given that the luminosity changes by a factor of ∼600 within the sample.

Surface brightness selection effects must be seriously considered in examining the correlations in Figure 7 (see, for example, earlier discussion by Impey & Bothun 1997, and references therein). Here, the work of Koposov et al. (2008) provides the most relevant reference. The selection functions are complex and depend on distance, magnitude, and surface brightness. However, Table 3 of Koposov et al. (2008) indicates that galaxies as faint as M_V ∼ −4.4, at distances of ∼180 kpc, could be recovered even if their central surface brightness is as faint as 29.9 mag arcsec⁻², more than 2 mag fainter than the median observed surface brightness at faint luminosities. There is also tentative evidence that the M31 satellites appear to behave in a similar way to the MW satellites (although they do not extend to the same low-luminosity limits). Searches for M31 satellites are subject to different (and as yet unquantified) selection effects compared to the MW satellites. Taken together, it appears possible that the "break" in the scaling relations at around M_V ∼ −9 may be real. If so, it may imply that—as a result of either formation (e.g., inflow of gas into dark matter haloes) or subsequent feedback/evolution (e.g., star formation)—there is a low-density limit to the central stellar densities of dwarf galaxies. If the effect is instead due to selection, then we are only observing those low-luminosity galaxies with the highest surface brightness, indicating that there may be very significant scatter in this property at faint magnitudes. Regardless of this speculation, the behavior of surface brightness with luminosity shown in Figure 7 certainly requires further consideration and explanation.

Table 4 lists various information regarding the masses and kinematics of the galaxy sample.