But it is not merely in quantity that our knowledge of dwarf galaxies has progressed in recent years. Our most basic perceptions of these objects has fundamentally changed. Back in 1971, there was very little information regarding the gaseous and stellar internal kinematics of dwarfs. However, he did note at that time that some dwarf spheroidal systems (dSph; most notably Ursa Minor, Hodge and Michie 1969) appear to have mass densities so low that it was hard to understand how they could have survived to the present epoch so close to the Galaxy. There was also very little information on the chemical properties of these galaxies. Of course, it was `obvious' to most astronomers at that time that the stellar populations of early-type dwarfs would be identical to that of globular clusters - after all, the nearest dwarfs were simply halo or inter-galactic systems that only happened to be structurally different from clusters. Dwarf irregulars, on the other hand, were known to be forming stars vigorously, but the full significance of this was not widely appreciated at the time (Searle et al. 1973). Even though Paul Hodge's review suggested otherwise, everyone knew that dwarfs `didn't matter' - these puny galaxies could not possibly play a significant role in large-scale issues of galactic evolution.
How things have changed! Now we know that dwarfs have undergone extraordinarily complex star-formation histories - basically, no two are alike (Mould and Aaronson 1983; Grebel 1997; Mateo 1998). Yet how these galaxies have repeatedly acquired, then lost the gas as needed to fuel this activity remains completely unknown. We can now measure the internal kinematics of dwarfs precisely using radio and optical techniques; these observations reveal that all dwarfs appear to contain dark matter (DM; Aaronson 1983; Mateo 1998)*. Chemical abundances have been determined for H II regions, planetary nebulae, and stars in many dwarfs, and these small galaxies now anchor our understanding of the chemical evolution of the Universe thanks to their low, nearly primordial, mean abundances (Skillman et al. 1989; Skillman et al. 1997; Mateo 1998). There is also growing evidence that dwarfs may play a significant role in the formation of larger galaxies (e.g. Mateo 1996; Unavane et al. 1996). The recent discovery of the Sgr dwarf (Ibata et al. 1994) caught in the act of dissolving into the Galactic halo beautifully illustrates how dwarfs may contribute to the formation of larger galactic systems. Finally, although in 1971 dwarfs seemed to be rare, we now recognize them as the most common type of galaxy; dwarf systems have now been identified in large numbers in a variety of environments within and beyond the LG (e.g. Karachentseva et al. 1985; Binggeli et al. 1985; Caldwell and Bothun 1989; Côté et al. 1997; Trentham 1998).
Figure 1 illustrates yet another remarkable case - the Sculptor dSph galaxy. Each of the four panels represents results from the individual 14.7' × 14.7' CCDs of the `Big-Throughput [a.k.a Bernstein-Tyson] Camera' (or BTC; http://www.astro.lsa.umich.edu/btc/tech.html) obtained in late September, 1998. Chip 4 was placed on the center of Sculptor, while Chips 1 and 3 were located on the Northern and Eastern edges of the galaxy, respectively. Chip 2 contains very few Sculptor stars - a somewhat surprising result I will return to later. The results for Chip 4 demonstrate that, as in other dSph systems, Sculptor has experienced numerous distinct episodes of star-forming activity. Clearly visible in Figure 1 is a dominant old population - probably somewhat older than 12 Gyr - corresponding to the dominant population observed by Da Costa (1984) in the initial CCD study of the galaxy. A little `spur' of stars is also visible that extends up from the principal turnoff region; this represents the turnoff of a younger population with an age of approximately 8 Gyr. Both populations feed stars into the well-defined sub-giant region, but the younger population systematically contributes bluer subgiants. This does not appear to be merely an age effect; rather, it seems that the younger stars are more metal-poor than the older stars in Sculptor! A considerably younger population is also present in the `blue straggler' region of the Chip 4 CMD. By comparing the number of stars in this region with the number in the same region in Chips 1 and 3, it becomes evident that these stars represent an additional turnoff component in Chip 4, and not merely additional blue stragglers. The existence of three turnoffs spanning an age range of at least 8 Gyr is no longer considered unusual in dSph galaxies; rather it has become an accepted - though still poorly understood - fact that most of these systems have experienced complex star-formation histories, and that in some cases, the galaxies may even be characterized by non-monotonic chemical enrichment histories!
Another interesting feature of Figure 1 is the clear gradient in the stellar population within Sculptor. Note that while Chip 4 exhibits the composite turnoffs described above and contains a rich population of red HB stars, neither feature is observed in Chips 1 or 3. Sculptor's complex (SFH) is confined to its central region. This suggests that the red HB must be intimately associated with the young turnoff(s), providing compelling evidence that age is the so-called second-parameter for the stellar populations in Sculptor.
The clear lack of Sculptor stars in Chip 2 is surprising. Our main-sequence counts allow us to detect the galaxy to an equivalent surface brightness of nearly 31-32 mag arcsec-2 (see below and Mateo et al. 1998a for details), yet it is clear that Sculptor only barely extends into the area covered by this CCD. According to the isopleth maps of Irwin and Hatzidimitriou (1995; hereafter IH), we should have seen Sculptor stars throughout this region. Apparently, either the smoothing adopted by IH blurred the edge of the galaxy or else their estimate of the background star/galaxy contamination was slightly underestimated. Regardless, the key point I want to stress here is that with large-area CCD detectors on large telescopes, it becomes possible to use deep, main-sequence star counts to trace the structure of these nearby dwarf systems in detail.
Because we now have reasonably good estimates of the SFHs of most Local Group galaxies, it has become possible to piece together the global star-formation history of our little corner of the Universe. Madau et al. (1998) have described an approach to do this for high-redshifts by producing the so-called `Madau diagram' - the star-formation rate as a function of look-back time - for distant galaxies. For their sample, the global star-formation history rises steadily with look-back time, peaking at intermediate ages in the range 5-7 Gyr in the past. Tolstoy (1998) has done the same analysis for local galaxies by combining the SFHs of individual systems within the Local Group. In contrast, she finds a global SFH that rises steadily towards 10 Gyr and older where the age resolution becomes relatively poor.
Both approaches have clear biases: studies of distant galaxies will be strongly influenced by high-surface brightness systems, both because such galaxies naturally stand out in any sample, and because of the strong cosmological decline in surface brightness (Σ ∝ (1+z)-4). For the local sample, we may simply not have obtained a fully representative collection of galaxies at the present epoch even when considering the entire Local Group. However, if by analogy to studies of large-scale structure, a `Stellar-Evolutionary Cosmological Principle' holds, there must be some volume surrounding us that contains a representative sample of galaxies that can be used to determine the SFH of the Local Universe. Determining the SFHs throughout this volume will be difficult and time-consuming; even with HST it remains challenging to identify the oldest main-sequence turnoffs in galaxies at the distance of M 31, and it is certainly beyond our present capabilities to do so outside the Local Group. There are many interesting dwarf targets now known in the M 81, Centaurus, and Sculptor groups (e.g. Karachentseva et al. 1985; Côté et al. 1997; Bremnes et al. 1998) that may one day help us better define the SFH of the Local Universe as NGST and large ground-based telescopes with adaptive optics come on line over the next decade.
How has this reversal of thought come about? The long history of the non-detection of an expected metallicity gradient among halo globular clusters (Searle and Zinn 1978), and the discovery of an age spread among globulars that is considerably longer than the free-fall time of the halo (VandenBerg et al. 1996) is well known to most astronomers. Over the past few years, many additional observations of field stars within the Galactic halo have provided a steadily growing body of circumstantial evidence that discrete structures populate the halo. For example, isolated clumps of stars with unusual kinematics have been identified serendipitously by many authors (e.g. Majewski 1992; Côté et al. 1993; Arnold and Gilmore 1992). Most spectacularly, the discovery by Ibata et al. (1994) of the Sgr dwarf proves that dwarf galaxies are being destroyed in the halo today. As we describe in detail below, the main body of Sgr represents only the core of a stream that extends at least 70° across the sky (Mateo et al. 1998a). The Sgr `dwarf' is now one of the apparently largest coherent structures in the sky! These new results suggest that the standard Eggen, Lynden-Bell and Sandage (1962) monolithic collapse picture probably cannot explain the formation of the entire halo. A number of large-scale studies (Majewski 1998; Harding et al. 1998; Beers 1998) are underway to try to identify coherent structures throughout the halo. Other observers have begun to detect low-surface brightness streams in the halos of more distant galaxies (Shang et al. 1998), and in the intracluster regions of the nearby galaxy clusters (Trentham and Mobasher 1998; Calcáneo-Roldán et al. 1998).
Since its discovery in 1994 (Ibata et al.), the known extent of the Sagittarius dwarf galaxy (Sgr) has grown steadily. The first map revealed a galaxy with a possibly clumpy structure subtending a region about 8° × 5° in size, and oriented roughly perpendicular to the Galactic plane. Subsequently, a series of papers reported a far larger projected size of at least 20° × 8° for Sgr (Mateo et al. 1996; Alard 1996; Fahlman et al. 1996; Ibata et al. 1997; Alcock et al. 1997) based on observations of various stellar tracers. Recently, Siegel et al. (1997) reported a possible detection of Sgr far beyond this 20° × 8° boundary.
Because Sgr is located only 16 kpc from the center of the Milky Way (Ibata et al 1994; Mateo et al. 1995), there is universal agreement that it is experiencing a strong tidal encounter. Generic simulations of dwarf-satellite destruction, as well as models specific to Sgr (Allen and Richstone 1988; Moore and Davis 1994; Piatek and Pryor 1995; Oh et al. 1995; Johnston et al. 1995; Velázquez and White 1995; Ibata et al. 1997; Zhao 1998) reveal that a strong tidal encounter will draw leading and trailing tidal streams out from the main body of Sgr during its closest encounters with the Milky Way. These streams should extend close to the projected major axis of the galaxy as stars migrate along orbital paths close to that of the disintegrating system. Evidence already exists for extra-tidal stars in some halo systems (globulars, Grillmair et al. 1995; dSph's, Kuhn et al. 1996, see discussion in Olszewski 1998).
Recently, we (Mateo, Olszewski and Morrison 1998) carried out a systematic search for a distinct tidal stream extending from the Sgr dwarf. Our goal was to identify main-sequence stars in Sgr along the extension of its major axis; the center of Sgr was assumed to coincide with the globular cluster M 54. The locations of the 17 SE major-axis fields that we ended up observing are shown in Figure 2. We also observed seven control fields located opposite of the l = 0° meridian from the corresponding target fields. Finally, we obtained data for four fields oriented perpendicular to the major axis of Sgr and passing through one of the major-axis fields; we refer to these as the `cross-cut' fields. The major-axis fields range from 10° to 34° from the center of Sgr.
The Sgr main sequence turnoff (MSTO) occurs at I0 ∼ 20 mag and (V-I) ∼ 0.6 (Mateo et al. 1996; Fahlman et al. 1996). By good fortune, this corresponds to a gap in the density of contaminating foreground stars from the thin disk (located at I0 ≤ 20 in this color range) and background galaxies fainter than I0 ∼ 22.5 at this color. To improve the contrast of Sgr with these contaminants, we therefore counted stars only within a box coinciding with the upper MSTO region in Sgr, after de-reddening the data using the extinction maps of Schlegel et al. (1998). Figure 3 shows the main-sequence counts along the Sgr major-axis and for the control fields after subtracting the contribution from the control fields. Even in the outermost Sgr point - located 34° from the galaxy's center! - there is a clear excess of stars relative to the controls. We were able to calibrate these counts in terms of surface brightness (SB) in units of mag arcsec-2 based on earlier results (Mateo et al. 1996); notice that in the outermost Sgr fields, where we detect a 4-6σ excess over the control field counts, the V-band SB is ∼30.5 mag arcsec-2. We looked for but did not detect evidence for a counter-stream to the NW of the center of Sgr; however, the large and rapidly variable reddening in these fields make it impossible to claim this as a significant negative detection.
These data have proven useful to improve estimates of the integrated luminosity of Sgr, and to probe the galaxy's structure. Perhaps the most interesting feature of the radial profile of Sgr is the kink at R ∼ 20°. Does this represent the transition from a dynamically distinct portion of the Sgr dwarf and a tidal stream - the `Sgr stream' - pulled out of the galaxy? Or are we seeing the current (inner) tidal stream in the process of joining an older, more extended stream from a previous tidal encounter (Zhao 1998)? One key to addressing these questions will be to define the orbital path of Sgr by determining distances along the stream, as attempted by Alcock et al. (1997), and to better define the projected distribution of Sgr stars on the sky. The only well-populated features in the Sgr CMD in these outer fields are the main sequence and sub-giant branch, neither of which are well suited for precise distance determinations. Nevertheless, we have unambiguously detected Sgr at an angular distance that implies overall dimensions of at least 68° × 8-10°, or about 30 × 4 kpc2. Since the projected scale length of the outer parts of Sgr is 17.2°, we should be able to use the techniques described here to trace the galaxy 20-30° further out than we have mapped it so far. Wide-field time series observations along the stream may reveal RR Lyr variables or dwarf Cepheids (Mateo et al. 1998b) that can be used to determine precise distances along the outer extension of the Sgr dwarf.
These techniques might also prove useful to (finally) constrain whether or not stars are associated with the Magellanic Stream (MagS). Since its discovery nearly 25 years ago (Mathewson et al. 1974), the MagS has steadfastly avoided a comprehensive explanation for its existence. The appearance (linear), size (over 100° long), and geometry (parallel to the motion of the Magellanic Clouds; Jones et al. 1994) of the Stream has long suggested a tidal origin (Murai and Fujimoto 1980; Lin et al. 1995; Gardiner and Noguchi 1996). However, all tidal models predict not only the ejection of gas, but of stars as well. The stream is currently observed to be entirely composed of H I clouds: no associated stars have yet been convincingly detected. The long history of optical searches for a stellar component of the stream has been described by Irwin (1991). Recently, Ostheimer et al. (1997) claim a possible identification of red giants associated with the stream, but their abstract is somewhat vague about the detection. Other searches, for example for C stars, may be doomed to a null result if the stellar component extracted to form the MagS did not happen to contain many of the tracer objects.
Two contending ideas have alternated in popularity over the years as an explanation for the MagS: the tidal models described above, and models invoking ram-pressure stripping of gas from the main body of the Magellanic Clouds (Moore and Davis 1994). Recently, Putman et al. (1998) have reported the identification of a leading Magellanic Stream. This adds considerable weight to a tidal interpretation of this feature and the classical trailing Stream, though, remarkably, ram-pressure models are not excluded (Sofue 1994; Moore and Davis 1994). If this really is a tidal feature, where are the stars?
One problem with identifying stars unambiguously in the MagS is that if they are uniformly distributed, their surface density is extremely low. For example, if as much mass was injected in stars into the MagS as in neutral gas (∼108 Msun) and these stars are as widely distributed as the gas in the sky (over an angular area in excess of 1000 deg2!), then the mean V-band stellar surface brightness is 28.3 mag arcsec-2 (we assume (M/L)V = 2.0). Obviously, standard techniques are inadequate to detect such a low surface density stellar feature. Can star counts similar to those employed for Sgr do the job? From Table 1 we can see that if we try to identify the red giants in the MagS using 2-color counts, we can expect at best a 3-sigma detection. Of course, the surface brightness could vary drastically over the sky if the stellar distribution is as clumpy as the gas. Although it is certainly possible to detect the stream in this manner, it is also clear that a more conclusive approach is desirable.
If instead of red giants, one tries to detect MS turnoff stars, the statistics become much more favorable. As noted in Table 1, two-color main-sequence star counts to V = 24.5 can detect a stream at the expected mean MagS surface brightness at a 15-sigma significance level. This has two important consequences. First, we can use this technique to map the stellar component of the stream. Second, and more importantly, we can use such counts to place strong limits on the total stellar content of the stream if no stars are found. For example, a failure to detect MS stars in the stream to an effective V-band surface brightness of ΣV = 30.5 mag arcsec-2 means that there are no more than about 107 Msun in stars, or about 10% of the H I mass, is in the MagS if the stars and gas are spatially coincident. But this too can be tested by obtaining deep counts over a large area - similar to the cross-cuts we employed in Sgr - across which the stellar component of the MagS may be found.
| ΣV | NRGB | S/N | NMS | S/N | ||
|---|---|---|---|---|---|---|
| 28.0 | 25 | (10) | 3.7 | 1250 | (1800) | 18 |
| 29.0 | 10 | 1.9 | 500 | 7.8 | ||
| 30.0 | 4 | 0.8 | 200 | 3.2 | ||
| LMC (D = 50-60 kpc; VRGB ≤ 18.0; VMS ≤ 24.5) | ||||||
Recently, we (Cook et al. 1998) found that the kinematics of the gas in the M 31 satellite LGS 3 is similar to that of its neutral gas (Young and Lo 1997b). It was already known that the spatial distribution of the gas and stars in this galaxy were also similar. However, if we consider systems that contain more obvious gas reservoirs, we find that in few cases do the stars and gas share the same spatial distribution and kinematics. For example, Leo A also has a symmetric H I structure centered on the optical component of the galaxy (Young and Lo 1996). However, the H I envelope of Leo A dominates its visible mass; it remains unknown if the optical kinematics are similar since these have not yet been measured. In most other cases, there is strong evidence for distinct stellar-gaseous dynamical configurations. For example, NGC 185 and NGC 205 both contain a neutral gas component, but in each case the asymmetric distribution and kinematics suggest that the gas is not in dynamical equilibrium (Young and Lo 1997a). Some of the gas in NGC 185 appears to be highly excited, implying a possible energy input from a recent supernova (Gallagher et al. 1984; Young and Lo 1997a). NGC 205 is clearly tidally disturbed by M 31, and seems to have experienced significant recent star formation near its core (Hodge 1973; Jones et al. 1996; Lee 1996); both effects could have input considerable energy into its ISM component. Other nearby early-type dwarfs may also possess neutral gas, specifically, Tucana, Phoenix and Sculptor (Oosterloo et al. 1996; Carignan et al. 1991; Young and Lo 1997b; Carignan et al. 1998). In each case, the putative ISM is distributed far from the galaxy centers, often asymmetrically. In Sculptor the H I is probably also not in dynamical equilibrium and may be moving radially relative to the galaxy center, though whether this is inflow or outflow is impossible to tell from the H I geometry alone. There seems to be no star formation associated with the Sculptor gas (Carignan and Demers, private communication; Figure 1). All of these galaxies exhibit a significant range in mass, SFH, and susceptibility to tidal effects (Mateo 1998).
I believe that the key to many of the mysteries associated with dwarf galaxies that I discuss in this review will be understood once we can unravel the gas-star puzzle in these systems. At the very least, we probably need to carefully reconsider our basic assumptions of how dwarf produce, lose, and, most intriguingly, re-acquire gas in the remote halo or intra-galactic regions of the Local Group. The recent paper by Blitz et al. (1998) may be an interesting starting point. More detailed dynamical models, especially ones that explicitly account for hydrodynamical effects in the gas, would be particularly useful to make some progress on this front (e.g. Mihos and Hernquist 1994; Hibbard and Mihos 1995).
The (M/L) ratios of all of the galaxies in Figure 4 can be adequately reproduced by the equation (M/L) = 1.5 + Mdark/L, where (M/L)s is taken as 1.5, the mass of the underlying dark halo, Mdark, is 2·107 Msun, and all quantities are in V-band solar units. The remarkable implication of this relation is that no dwarf has a mass smaller than Mdark ∼ 2.0·107 Msun. Moreover, if the distribution of baryonic mass in these galaxies extends to even lower luminosities than depicted in Figure 4, there should be a number of essentially `dark' systems floating around in the Galactic halo.
Many people have reviewed the `standard' DM interpretation for the unusual kinematics of dwarf systems in recent years (Pryor 1996; Mateo 1997; Olszewski 1998; Mateo 1998). However, every now and then I like to remind myself that the very term `dark matter' already implies a solution to what is really a fundamental problem of galactic kinematics. I will close by discussing two possible alternatives to DM as they apply to nearby dwarfs.
Tides Kuhn and Miller (1989), Kuhn (1993), Kroupa (1997), and Klessen and Kroupa (1998) have argued that tidal effects can induce resonances in dSph systems that can artificially inflate their central velocity dispersions, hence mimicking dark matter. Pryor (1996) has argued that such heating is difficult to understand as an explanation for the large central velocity dispersions observed in dSph galaxies. Detailed n-body simulations of encounters of dSph galaxies and the Milky Way at impact parameters ranging from 10-50 kpc (Piatek and Pryor 1995; Oh et al. 1995; Johnston et al. 1995) suggest that close encounters can produce streaming motions in outer parts of the dwarfs. These motions will be seen as a systematic change of the mean velocity along the major axis and can be incorrectly interpreted as rotation. However, despite this streaming in the outer regions of the dwarfs, such close encounters negligibly alter the central velocity dispersion. Although strong evidence for tidal extensions and perturbations in some of the nearby dSph systems is accumulating rapidly (Mateo et al. 1996; Alard 1996; Fahlman et al. 1996; Kuhn et al. 1996; Kleyna et al. 1998), there are no convincing demonstrations yet of strong tidal heating of the core of a dSph galaxy, apart possibly from the Sgr system which appears to be in the terminal stages of tidal disruption as it passes extremely close to the Milky Way (Bellazzini et al. 1996; Mateo 1998; Mateo et al. 1998a; though see Ibata et al. 1997).
Two recent observations further help rule out tides as a significant contributor to the dark matter problem. In the case of Leo I (Mateo et al. 1998b), the central dispersion and inferred central (M/L) ratio of Leo I is difficult to understand in a tidal model given the remote location and large systemic velocity of this system relative to the Milky Way. Byrd et al. (1994) modeled the orbit of Leo I within the Local Group over a Hubble time and found that it would have only once passed close to a large galaxy - the Milky Way - some 2-4 Gyr ago. These results are completely at odds with any resonance heating mechanism: In the models by Kuhn and Miller (1989) and Klessen and Kroupa (1998), significant core heating does not occur until a galaxy has experienced several orbits around the Galaxy on timescales ∼1 Gyr. Cook et al. (1998) found that the isolated dwarf LGS 3, a remote companion of either M 31 or M 33, has a `normal' - that is, high - (M/L) ratio. Although there are no detailed models of plausible orbits for this system, it represents another example of a relatively isolated dwarf that appears to require dark matter despite little opportunity for tides to have externally heated the system.
Kinematic studies of other isolated dwarfs such as Antlia and Tucana would be particular helpful in definitively settling this issue and to better understand the kinematics of dwarf galaxies and streams that clearly are affected or originate from tidal encounters (e.g. Sgr). Precise optical kinematic studies of such remote galaxies will soon be feasible as the new-generation 6.5-8 m class telescopes come on line. On balance, however, the present data suggest that tides do not play a significant role in heating the cores of remote galaxies such as Leo I and possibly LGS 3.
Modified Gravity Milgrom (1983a,b) introduced Modified Newtonian Dynamics, or MOND, to understand the rotation curves of disk galaxies (and some other related phenomena) with a modified form of Newton's law of gravity without resorting to DM. Only at very low accelerations (defined by the parameter a0 ∼ 1.5·10-8 cm sec-2) is Newton's law substantially altered from its standard form (McGaugh and de Blok 1998). What is so nice about MOND is that a single unambiguous example of a galaxy strictly obeying Newtonian dynamics in the low-acceleration regime would falsify the theory instantly. How does MOND fare with LG dwarfs?*
Lake and Skillman (1989) and Lake (1989) suggested that the rotation curves of IC 1613 and NGC 3109 could not be explained by MOND unless a0 ≤ 3·10-9 cm sec-2, a value incompatible with that needed to interpret rotation curves of giant systems. Milgrom (1991) noted that (a) MOND successfully fit the shapes of the rotations curves of these and other galaxies discussed by Lake (1989), and (b) the mixed success that MOND had in reproducing the amplitudes of the rotation curves could be understood given the errors in the galaxy distances, inclinations, and asymmetric-drift corrections. For NGC 3109 (Jobin and Carignan 1990), Milgrom was in fact justified in claiming the earlier rotation curve was in error, though in the case of IC 1613 it remains unclear if Milgrom's objection to Lake & Skillman's (1989) conclusion is valid. More recently, Sanders (1996) found that the rotation curves of both NGC 55 and NGC 3109 are fit well by MOND, while McGaugh and de Blok (1998) find that MOND does a good job of accounting for the kinematics of low surface-brightness disk galaxies.
Gerhard and Spergel (1992) argued that the internal kinematics of LG dSph galaxies demanded DM even if MOND was used to estimate their masses. Lo et al (1993) found that the MOND masses for many nearby dIrr galaxies were smaller than their integrated H I masses - an obvious failure if correct. Milgrom (1995) responded that if the observational errors and most recent results were considered, neither effect claimed by Gerhard and Spergel (1992) was observed. In the second case, Lo et al (1993) used an incorrect expression to determine the MOND masses, leading to estimates that were too small by a factor of 20 (Milgrom 1994); when the correct expression is used, the MOND masses are consistent with the inferred luminous masses of the galaxies without the need for a dark component.
Very recently, Anderson et al. (1998) identified anomalies in the motions of the Pioneer and Ulysses spacecraft currently located in remote regions of the Solar System and beyond. Is this an empirical measure of possible non-Newtonian gravitational effects? Or simply an extremely subtle, unaccounted acceleration from the spacecraft components (Katz 1998)? The burden of proof remains squarely on MOND, but the kinematic data for LG dwarfs does not yet refute this alternative to DM. It is particularly exciting to realize that we may be finally approaching a time where MOND and other non-Newtonian predictions can be actively tested on solar-system scales.
| First version: | 10th | October, | 1998 |
| Last update: | 16th | November, | 1998 |