Reprinted from Seventeenth Texas Symposium on Relativistic Astrophysics and Cosmology

Volume 759 of the Annals of the New York Academy of Sciences

September 30, 1995

THE GALACTIC CENTER

R.Genzel, A.Eckart and A.Krabbe

Max-Planck Institut für extraterrestrische Physik Garching, FRG

ABSTRACT

The Galactic center is a superb laboratory of modern astrophysics where astronomers can study at unprecedented spatial resolution and across the entire electromagnetic spectrum physical processes that may also happen at the cores of other galaxies. Following a brief review of the phenomena observed in the central few hundred parsecs, we discuss two main issues in more detail. The source(s) of luminosity in the central parsec and the nature of the compact radio source SgrA*, the most likely candidate for a massive central black hole.

I. INTRODUCING THE PHENOMENA

Until about 20 years ago, investigations of the Galactic nucleus were severely impeded by the large amount of interstellar dust and gas situated in the Galactic plane between the Sun and the nucleus, and preventing observations in the visible, ultraviolet and soft X-ray bands. In the past two decades, however, our knowledge about the Galactic nucleus has improved dramatically, as sensitive high resolution observations in the radio, infrared, hard X-ray and gamma-ray bands have become available. In the following, a brief summary of these phenomena as revealed by a number of observations is given, followed by a more detailed account of recent results on the exploration of the central parsec. For more extensive discussions we refer to recent reviews (1,2,3).

The nuclear mass is dominated by stars, except perhaps in the innermost parsec. Infrared observations on a scale of 100 to 1000 pc show that these stars appear to be distributed in a rotating bar (4). Binney et al.(5) have pointed out that the gravitational torque of this bar may account for the non-circular motions of interstellar gas clouds found by radio spectroscopy (6,7). The non-circular motions in turn may trigger gas infall into the nucleus (5). X-ray spectroscopy of 6.7keV Fe K-shell emission (and recently of other K- and L-shell lines) has revealed a component of ~108 K intercloud, coronal gas that permeates the central bulge/disk on a scale of about 100 pc (8,9). The coronal medium cannot be gravitationally bound to the nuclear bulge and probably escapes as a Galactic wind. There is increasing evidence from gamma-ray spectroscopy of the 1.8 MeV 28Al line (10) and from infrared stellar spectrophotometry (11,12, see below) that (massive) star formation has occured throughout the Galactic center region no longer than 10 million years ago. Hence the energy input responsible for the coronal gas could be from supernovae that have been exploding in the central bulge (3). Also on a scale of 100 pc several variable, spectacular hard X-ray and gamma-ray sources have been found (13). They may represent stellar black holes or neutron stars accreting gas from a companion or from nearby dense gas clouds. A broad emission bump at ~500 keV and a twin radio jet seen in one of these sources (the "Great Annihilator' 1E1740.7-2942) may signify the presence of a relativistic electron-positron jet annihilating in the environment of the compact source (14,15,16). Eisewhere in these proceedings, Mirabel (see also 17) gives a detailed account of the recent detection by radio interferometry of superluminal motions of radio knots in another hard X-ray/gamma-ray source near the center, providing direct evidence for the above interpretation of relativistic radio jets in the nuclear region of our Galaxy.

Throughout the central few hundred parsecs giant molecular clouds (105.5-107 Mo, N (H2)~1023.5 to 1024 cm-2) are found whose gas density (n(H2)~104 to 106 cm-3) and temperature (40 to 200 K) are significantly greater than those of the clouds in the Galactic disk (18). The dust temperature of these clouds, on the other hand, is fairly low indicating that most of them are presently not heated intemally by active star formation. The dynamics of this central molecular cloud layer is characterized by large intemal random motions and unusual streaming velocities that can be partially explained by the presence of the central bar potential (18, 19, 20). Extended 6.4 keV line emission from 'neutral' iron (9, 21) and 8.5-22 keV hard X-ray continuum emission (22) appears to be correlated with these very dense molecular clouds. This surprising finding suggests that the molecular clouds act as dense reflectors of hard X-ray emission originating somewhere in the central 100 pc, perhaps analoguous to what is observed in some active galactic nuclei (23). The origin of the scattered X-ray emission is yet unkown. It could be identical with the known compact sources discussed above (22), or perhaps with an active source at the very center (9).

Magnetic fields as large as ~1 mGauss appear to permeate the central 50pc and are aligned approximately perpendicular to the Galactic plane (24). Where they interact with neutral gas clouds spectacular filaments of nonthermal radio synchrotron emission are seen. The most spectacular such set of filaments, the Radio Arc, located about 13' (34 pc) north of the dynamic center and the central radio source SgrA also contains a set of more than 20pc long filaments of thermal radio emission that are associated with ionized plasma of temperature about 8000 K (25). The origin of the widespread ionization, whether caused by collisions of fast moving clouds, by magneto-hydrodynamic effects or by photoionization has been the subject of a lively debate for some time (26- 29), with the odds recently turning in favor of photoionization by a number of OB associations (11, 3). At radii between 1.5 and 5pc from the dynamic center there is a system of orbiting molecular filaments approximately arranged in form of a circum-nuclear 'ring' or 'disk' (30). This circum- nuclear disk (CND) is probably fed by gas infall from dense molecular clouds at >10pc (31) and appears to drop filaments or streamers into the central 1.5pc which is comparatively devoid of interstellar matter (the "central cavity", 32). The ultraviolet (UV) radiation in this cavity is sufficiently high to ionize and heat to about 6000 K a significant fraction of these streamers. The central ionized streamers orbit the dynamic center (33, 34) and are arranged in form of a "mini-spira1' that comprises the brightest part of the central thermal radio source SgrA (West). While the streamer velocities are largely dominated by the gravitational field (34, 35), an intense nuclear wind (mostly from massive stars discussed below) probably also affects their motions within about 1 pc from the center (36,37, 39). Infrared polarimetry and radio observations of Zeeman splitting indicate that the streamers are permeated by mG-magnetic fields that are dragged along their orbits (40-43). The magnetic fields may account for macroscopic viscosity and angular momentum transport in the clumpy, turbulent circum-nuclear gas streamers. The eastern part of SgrA is a synchrotron shell source that appears to be expanding into and accelerating dense molecular gas (44, 29, 45-46 ) . SgrA (East) thus may be direct evidence for a recent explosion (t < 105 years) in the central 10 parsecs, with an energy of 1052±1 ergs (29,45-46).

The density of the nuclear star cluster increases with decreasing radius R approximately as R-2 outside of its core radius of less than a parsec in size (47,48). Inside that core radius ~ the stellar density is certainly a few 106 (47,48) and possibly several 107 times greater (49) than in the solar neighborhood. Direct collisions and mergers between stars and collisional destruction of stellar envelopes may thus play a role there (50,3). Infrared spectroscopy has identified individual blue and red supergiants in the core (51-53). These supergiants are very likely massive stars that have formed in the core within the last few million years. They probably provide a significant and perhaps dominant fraction of the total luminosity of the central few parsecs. A discussion of the present state of our knowledge about this nuclear cluster, its distribution and evolution will be given in the next section. Also inside that radius, the stellar and gas velocities are observed to increase, which may signal the presence of a black hole with mass 1 to 3x106 Mo, at the dynamic center (34,1). A compact radio source, SgrA*, lies approximately at that position and is close to, but not coincident with a bright near-infrared source (IRS16) of blue color. SgrA* has stellar dimensions (size a few AU, 54), but is presently not very conspicuous in any wavelength range other than the radio range. This is somewhat surprising if SgrA* is in fact a massive black hole surrounded by stars and gas clouds.

Clearly the modern multi-wavelength observations of the Galactic nucleus tell a fascinating story and show that a broad range of phenomena involving a number of physical processes are at work. To help the reader find the way in the jargon of names Fig.1 gives a "road map" of the central 100pc.

II. WHAT POWERS THE CENTRAL PARSEC?

Fig.2 shows a composite spectral energy distribution of the central few parsecs (from 3). Strong 20 to 300µm continuum emission from 50 to 100 K dust grains originates in the circum-nuclear disk and in a cloud ridge associated with the northem arm of the mini-spiral (55, 32). The far- infrared emission can be used as a calorimeter for estimating the total short-wavelength (visible and UV) luminosity of the central parsec that cannot be accounted for by the main sequence and late type stars in the nuclear stellar cluster. Taking into account the (~50%) fraction of the nuclear radiation not intercepted by the CND and mini-spiral the total UV and visible luminosity of the central parsec has been estimated to lie between 1 and 2x107 L (55, 56). The Lyman continuum flux is about 2 to 3x1050 s-1 (~1.3 to 1.9x106 L0) as determined from the thermal radio continuum (44).

Infrared spectroscopy of the ionized gas demonstrates that SgrA (West) is a low excitation HII region with electron temperature ~6000 K and a range of electron densities (103 to 105 cm-3, 57-59). Combining the line ratios of the different fine structure lines, the flux density of the radio continuum and the fluxes of the hydrogen and helium radio recombination lines in an excitation analysis, a relatively low effective temperature (~35,000 K) is derived for the ionizing radiation field (57, 58, 60, Fig.2). Effective temperatures significantly in excess of 40,000 K are only possible if the strong 12.8µm [NeII] line and most of the radio continuum come from very dense gas (106 to 107 cm-3, 58), a solution that is not entirely excluded but not supported by the data. More detailed spectroscopy of species with higher ionization potentials and of density sensitive line ratios, planned with the spectrometers onboard the Infrared Space Observatory (ISO), should give a clearcut answer. The line ratios also suggest a metal abundance of about twice solar (57, 61, 62, 58). What powers this low excitation Hll region with a bolometric luminosity of 107±0.3Lo?

At this point we need to tum our attention to a number of recent high resolution near-infrared observations that pertain to the distribution and characteristics of the nuclear stellar cluster. These measurements have become possible through the advent in the last half a dozen years of sensitive, large format detector arrays. In only a few years time it has become possible to first record high quality seeing limited images (~1", e.g. 63) and then further to improve through speckle imaging the resolution to the diffraction limit of 4m class telescopes (~0.15" at 2µm, 49, 64). The presently best such 2µm image of the central parsec is shown in Fig.3 (adapted from 65) and was taken with the MPE SHARP camera on the 3.5m ESO NTT during superb (~0.3") seeing and resolves the near-infrared emission of the central parsec (23") into about 700 stars with K-band (2.2µm) magnitudes <16. The central IRS16 complex located within 1" of the compact radio source SgrA* consists of about two dozen single (or perhaps multiple) stars. From the number distribution of the near-infrared sources Eckart et al.(49, 65) conclude that the centroid of the stellar cluster is more likely on SgrA* than on the IRS16 complex and that the core radius of the K<14 2µm stellar number density distribution is about 0.2pc. If the stars with K<14 are representative of the overall mass distribution of the cluster (an assumption that still needs to be proven by spectroscopic identification of the stars) such a small core radius would imply that the stellar density in the core is in excess of 107.5 Mo pc-3. At these densities stellar mergers and disruption of giant star atmospheres by direct stellar collisions become very important, with a number of interesting consequences for stellar and cluster evolution (50, 66, 67, 3).

Another important ingredient of the near-infrared story has been the discovery by Allen et al.(52) and Forrest et al.(68) of a 2.06µm HeI/2.16µm Br gamma near-infrared emission line star (the AF-star), followed by the discovery of an entire cluster of about 15 such stars in the central parsec and centered on the IRS16 complex (53, 69). In fact recent subarcsecond HeI line imaging (53, 69, 65) and imaging spectroscopy with the new MPE 3D-spectrometer (Fig.4, 69) now unambiguously show that several of the brightest members of the IRS16 complex are HeI-stars, as is the nearby bright source IRS 13 (see also 70). The HeI "broad line region" discovered a decade ago by Hall et al. (71, see also 72) is now identified as a group of mass losing, luminous He-rich stars. Non-LTE stellar atmosphere modeling of the observed emission characteristics of the AF-star (73) confirms and quantifies the earlier conclusion (52) that the AF-star is similar to WN9/Ofpe stars, a rare class of evolved, luminous blue supergiants related to Luminous Blue Variables and Wolf-Rayet stars. According to the analysis in (73), the AF-star has a luminosity of about 105.5 Lo, effective temperature near 20,000 K and ZAMS mass between 25 and 40 Mo. The AF-star has a surface He/H abundance ratio near unity and loses mass at a velocity of 700 km/s and rate of 6x10-5 Mo yr-1. Based on the most recent results from 3D (Fig.4) a preliminary analysis of the brightest Hel stars (IRS16NE,C,SW, IRS 13) suggests that these stars may in most respects be similar to the AF-star, but about 5 to 10 times more luminous (Najarro, priv.comm.). Of particular interest is the spectrum of IRS 13 that shows 2.19µm HeII emission in addition to the very bright and broad wing HeI lines (Fig.4, see also 70). This finding implies that IRS 13 is significantly hotter than most of the other HeI stars (~30,000 K) and very luminous (several 106 Lo, Najarro, priv.comm.). IRS 13 and the IRS16 HeI stars appear to have a very large He abundance ({He}/{H}»1, 74) and thus are probably very late type Wolf-Rayet stars. They probably each emit several 1049 Lyman continuum photons per second. Finally Blum et al. (75) and Krabbe et al. (69) have also found the first evidence for classical WC (9) Wolf-Rayet stars, further strengthening the interpretation of the Hel stars as evolved massive stars. Combining the contributions from all its members, the HeI-star cluster can plausibly account for most of the bolometric luminosity (~1 to 2x107 Lo) and a significant fraction of the Lyman-continuum luminosity (~106.3 Lo) of the central parsec. The HeI-star cluster also provides in excess of 1038 erg/s in mechanical wind luminosity which may have a significant impact on the gas dynamics in the central parsec. In fact the outward acceleration of the winds may explain why a possible central massive black hole is not likely to accrete very much interstellar gas at the present time (3).

In agreement with earlier proposals by Rieke and Lebofsky (76) and Allen and Sanders (77), the recent stellar infrared spectroscopy in the central parsec perhaps is best explained by a star formation burst 5 to 7x106 ago in which a few hundred OB stars and perhaps a total of a few thousand stars were formed (3, 78, 69). The HeI stars may be the most massive cluster members that in the mean time have evolved off the main sequence. In this scenario the central parsec is now in the late, wind- dominated (30 Doradus in LMC!) phase of the burst. This model accounts naturally for the low excitation of the SgrA (West) HII region. There are several difficulties, however. First the present gas density in the central parsec is too low for conventional gravitational collapse of gas clouds to stars in the presence of the very strong tidal forces (79). Perhaps the burst was triggered by a large influx of dense gas less than 10 million years ago. Second if the Hel stars are in fact closely related to LBVs, conventional wisdom would give an estimate of the duration of that phase of less than 105 years. Consequently, either there would have to be 106.7/(<105)x15 >102.7 additional OB stars lurking in the cluster (if star formation has been semi-continuous) that have not been identified yet. Such a large number of OB stars, however, is inconsistent with the properties of the bright near-infrared sources (69, 65) and the far-infrared luminosity constraint. Alternatively, the burst would have to have been very sharp in time. In that case we would happen to observe this last nuclear burst in the LBV phase (with an a priori probability of less than 1%). The latter explanation, however, cannot explain the simultaneous presence of stars in the HeI-phase that are different by a large factor in luminosity (AF and IRS 13, IRS16 NE,C,SW), and hence in mass.

Based on earlier theoretical work by Lee (80) Eckart et al.(49) have proposed sequential merging as an alternative to the starburst scenario, a possibility whose likelihood strongly depends on a very high density of the nuclear cluster (3). In a recent Fokker-Planck calculation of an evolving Galactic center type, dense cluster with merging Lee (81), however, has found that there are even not enough 20 Mo stars and no >30 Mo stars as a core of density 107.5 Mopc-3 or greater cannot be maintained for a long enough time. Morris (79) has suggested that the HeI stars are not classical blue supergiants at all but objects that have been created in collisions between (~10 Mo) stellar black holes and red giants. Both accounts of the HeI stars just cited are very specific to the high density environment of the central parsec. But a number of stars that look just like the SgrA HeI stars have now been found in several clusters 2 to 13' away from the central, high density SgrA region (11). In the case of the Morris (79) scenario one would also probably expect a much larger X-ray emission as a result of the collisions than is observed (81).

In conclusion then of this chapter we think that a moderate star formation burst 5 to 7x106 years ago is in fact the most likely interpretation when coupled with the additional proposal that the 'HeI phase' in the evolution of a metal rich cluster is much longer than the traditional estimates of the LBV phase. It may comprise a significant fraction of the main sequence lifetime of massive stars. Such a proposal is consistent with recent stellar evolution work. Meynet et al. (82) and Langer et al. (83) propose that in high metallicity stars core nuclear synthesis products are dredged up earlier and mass loss rates are higher than previously thought. We finally note that there is other evidence that star formation in the Galactic center has been time variable. The fact that there are only a few red supergiants (L>104.5 Lo: IRS7, IRS12 N, IRS 14N) when compared to the number of blue supergiants suggests that there was relatively little star formation before 10 million years ago. Yet there a number of late type stars with luminosities 103 to 104 Lo, both inside (65, 69) and outside (84, 12) the central parsec. These moderate luminosity stars are likely asymptotic giant branch stars of moderate mass (2 to 8 Mo) that may signify another star burst episode ~108 years ago. We are convinced that the high quality near-infrared imaging spectroscopy will now permit within the next year a much better understanding of the evolution of the nuclear cluster and the massive stars it contains.

III. Is SgrA* a Massive Black Hole?

The next key issue that we want to discuss is the evidence for a central massive black hole. Ever since the original discovery of the nonthermal compact radio source SgrA* at the core of the nuclear star cluster (85-87) that source has been the primary black hole candidate, in analogy to compact nuclear radio sources in other nearby normal galaxies (88, 89).

In fact ever more detailed radio observations have confirmed the unique nature of SgrA* in the Galaxy. Recent very long baseline interferometry (VLBI) observations at 7mm show its size to be less than a few AU (54, 90). Its proper motion relative to a background quasar is now known to be less than about 38 km/s (54), almost 5 times smaller than the (2d-) velocity dispersion of the stars. It is therefore almost inevitable that SgrA* is a massive object with a mass in excess of about 100 Mo. The radio observations (54) have also provided compelling evidence for intrinsic flux density variability on time scales of months/years. The source shows a mm/submm excess above the flat cm-spectral energy distribution (91, Fig.2) probably indicative of the presence of a very compact (~1012 cm) radio core of stellar dimensions.

Yet observations at shorter wavelengths show nothing particularly impressive toward the position of SgrA*. The high resolution maps of Eckart et al. (49, 64) for the first time did show that there is near-infrared emission toward SgrA*, within the present ±0.2" relative positional uncertainty of the infrared/radio frames. However, the most recent data suggest that the near-infrared emission is most likely due to a local concentration of bright stars near SgrA*(65). The 2µm maps displayed in Fig.3 clearly indicate that SgrA*(IR) is resolved into about half a dozen compact sources. 2µm speckle polarimetry shows that the polarizations of these sources, with the exception of one knot ~1" south-east of the nominal position of SgrA*, are small and consistent with the dichroic absorption of magnetically aligned dust grains in the interstellar medium along the line of sight to the Galactic center (65). SgrA*(IR) also does not show intrinsic variability on scales of minutes or years, or significant line emission (65). Any one of the knots in Fig.3 that might be the true infrared counterpart of SgrA* has an absolute K-magnitude of about -3, similar to an early B main sequence star (~104.2 Lo) or an early K giant (L~102.2 Lo) .

(Variable) X-ray emission is usually considered a key signature of black holes. However, in contrast to the fairly bright infrared emission the X-ray luminosity of SgrA (West) is fairly low (Fig.2). The <2.5 keV X-ray luminosity (corrected for interstellar extinction) is only a few Lo (92, 93) and the hard X-ray luminosity (2.5 to 100 keV) is less than a few hundred Lo (94, 14, 95). The upper limit originates from the recent finding by ASCA that there is a compact hard X-ray source about 1' south-west of SgrA* that may have contributed significantly to some of the lower resolution hard X-ray observations just mentioned (9). The inevitable conclusion is that any central active source (SgrA* ?) presently contributes only a small fraction of the bolometric luminosity of the central parsec, with a conservative upper limit of perhaps a few 105 Lo (3, 96). There are currently three possible constraints on any past activity of SgrA*. The first two are based on the scattering of hard X-ray emission from SgrA*. Considering the extended 8.5 to 22 keV emission measured with the ART-P telescope on GRANAT, Sunyaev, Markevitch and Pavlinsky (22) conclude that 'if there is a supermassive black hole in the Galactic nucleus, it has not emitted at the level of its Eddington luminosity for even a day over the past 400 yr'. A somewhat more conservative calculation (3) shows that the average 8-22 keV luminosity of SgrA* was not larger than 105.2 Lo during the past 400 years (the Eddington luminosity of a 106 Mo black hole is ~1010 Lo). Extrapolating from the extended 6.4 keV Fe-line emission, Koyama (9) estimates that the hard X- ray luminosity of a central source (SgrA* ?) is on average larger than 105.4 Lo. The third possible constraint comes from the thermal energy of the 108 K gas that ASCA observes to be associated with SgrA(East) which may be associated with an energetic recent explosion (>1052±1 ergs, see above) from the nucleus.

The evidence for a central mass concentration in the Galactic center, perhaps in form of a massive black hole, thus is based entirely on observations of the gas and stellar dynamics. The presently available measurements, most recently reviewed by Genzel, Hollenbach and Townes (3), are summarized in Fig.5. It shows the enclosed mass as function of radius. The various mass estimates derived from the gas/stellar dynamics are compared to the mass distribution derived from the stellar light and a constant mass to light ratio (M/L~1 Mo/Lo, dashed in Fig.5). Compared to the situation more than a decade ago when the first velocity measurements of ionized gas clouds became available (57, 60) and also 6 years ago when the case was last reviewed at a Texas symposium (97) the evidence for a central dark mass of 1 to 3x106 Mo has become substantially stronger. Numerous independent gas and stellar dynamics measurements within the central parsec are now available, agree reasonably well and suggest an excess mass within ~0.5 pc above what accounted for by the stellar light. The excess indicated by any single one of the data points, however, is only at the 2 to 3.5 sigma level. Only when all measurements are taken together is the evidence for a central mass distribution quite compelling. Of greatest importance for this conclusion are the three data points at the innermost radius (0.2pc or 5") sampled so far. The large cross indicates the mass derived from the velocity field of the central part of the mini-spiral (35, 98, 99). The open circle denotes the mass derived from the stellar velocity dispersion of late type stars sampled in their 2.3µm CO overtone absorption feature (~125 km/s: 100, 101). The filled circle shows the mass derived from the velocity dispersion of the HeI stars (~135 km/s: 102). Taken at face value the density of the dark mass indicated by the data in Fig.5 is at least 108 Mo pc-3 with a mass to luminosity ratio of at least 10 Mo/Lo.

While the available dynamical measurements strongly point toward the existence of a 1 to 3x106 Mo black hole they are clearly not sufficient as a proof. The density of the dark mass referred to above is only a factor of two or so greater than the largest value for the core density of the stellar cluster (49). It is already clear, however, that the dark mass cannot be due to a concentration of old, low luminosity and low mass stellar remnants (e.g. neutron stars). Such low mass remnants could not have a core radius smaller than the massive HeI-stars or the intermediate mass late type stars sampled with the CO absorption feature. A possible configuration without a central massive black hole may be a central cluster of >10 Mo stellar black holes (79) of which SgrA* perhaps may be one member. Clearly the key for distinguishing between the case of a single massive hole and a concentration of massive stellar remnants will be to determine the dynamic mass significantly inside of 0.2 pc. Two experiments are now well underway in our group at MPE to a) determine with 3D the radial motions of emission/absorption line stars between 1" and 5" from SgrA* and b), measure the proper motions of all brighter stars within <2" (0.08 pc) of SgrA* from repeated high resolution near-infrared imaging with the SHARP camera on the ESO NTT. First results from a 3D data set taken in summer 1994 and from the first four epochs of the proper experiment (covering ~2.5 years) are very encouraging and promise to confirm (or disprove) a 1 to 3x106 Mo black hole in an about 6 year measurement period.

Adopting now the notion that SgrA* is indeed a million solar mass black hole, the riddle remains why it is presently so inactive. It is very interesting that the Galactic center shares this 'luminosity deficiency' or 'blackness' problem with essentially all nearby nuclei for which there is substantial evidence for dark central masses (103), including the perhaps best case, the 'super' H2O maser source NGC4258 (104). It is possible that the tidal disruption and accretion of stars by the hole (happening in the Galactic center at a rate of ~10-4 yr-1) occurs very efficiently albeit at low duty cycle (105). Accretion of interstellar gas streamers by the hole may be prevented by the need to overcome the angular momentum problem, coupled with the outward force of the stellar winds as discussed above. Finally, the wind gas itself may be accreted largely spherically, with very low radiation efficiency (106). Nevertheless current models of black hole accretion have to be stretched to be comensurate with SgrA* being an underfed million solar mass black hole (107).

Acknowledgement

We are grateful to our MPE colleagues M.Cameron, S.Drapatz, R.Hofmann, H.Kroker, D.Lutz, B.Sams, L.Tacconi-Garman, N.Thatte and L.Weitzel, as well as F.Najarro for letting us report 3D, SHARP and stellar modelling results prior to publication. We also thank K.Koyama and Y. Tanaka for access to unpublished ASCA results.

REFERENCES

  1. Genzel, R. and Townes, C.H. 1987, Ann.Rev.Astr.Ap. 25, 377
  2. Blitz, L., Binney, J., Lo, K.Y., Bally, J. and Ho, P.T.P. 1993, NATURE 361, 417
  3. Genzel, R., Hollenbach, D. and Townes, C.H. 1994, Rep.Progr.Phys. 57, 417
  4. Blitz, L. and Spergel, D.N. 1991, Ap.J. 379, 631
  5. Binney, J.J., Gerhard, O.E., Stark, A.A., Bally, J. and Uchida, K.A. 1991, MNRAS 252, 210
  6. Liszt, H.S. and Burton, W.B. 1980, Ap.J. 236, 779
  7. Bally, J., Stark, A.A., Wilson, R.W. and Henkel, C. 1987, Ap.J.(Suppl.) 65, 13
  8. Koyama,K. et al. 1989, NATURE 339, 603
  9. Koyama, K. 1994, preprint
  10. Diehl, R. et al. 1993, Astr.Ap.(Suppl.) 97, 181
  11. Cotera, A.S. et al. 1994, in (38),217
  12. Glass, I. 1994, in (38), 209
  13. Skinner, G.K. 1993, Astr.Ap.(Suppl.) 97, 149
  14. Sunyaev, R. et al. 1991, Ap.J. 383, L49
  15. Mirabel, I.F., Rodriguez, L.F., Cordier, B., Paul, J. and Lebrun, F. 1992, NATURE 358, 215
  16. Ramaty, R., Leventhal, M, Chan, K.W. and Lingenfelter, R.E. 1992, Ap.J. 392, L63
  17. Mirabel, I.F. and Rodriguez. 1994, NATURE 371,46
  18. Gusten 1989, in (112), 89
  19. Bally, J., Stark, A.A., Wilson, R.W. and Henkel, C. 1988, Ap.J. 324, 223
  20. Binney, J.J. 1994, in (38), 75
  21. Tanaka, Y. et al. 1994, in prep.
  22. Sunyaev, R.A., Markevitch, M. and Pavlinsky, M. 1993, Ap.J. 407, 606
  23. Awaki, H., Koyama, K., Inoue, H. and Halpern, J.P. 1991, PASJ 43, 195
  24. Morris, M. 1990,in Galactic and Extragalactic Magnetic Fields, eds. R.Beck, P.Kronberg and R.Wielebinski (Dordrecht:Kluwer), 361
  25. Sofue, Y. 1994, in (38), 43
  26. Serabyn, E. and Gusten, R. 1987, Astr.Ap. 184, 133
  27. Heyvaerts, J., Norman, C. and Pudritz, R. 1988, Ap.J. 330, 718
  28. Morris, M. and Yusef-Zadeh, F. 1989, Ap.J. 343, 703
  29. Genzel, R. et al. 1990, Ap.J. 356, 160
  30. Genzel, R. 1989, in (112), 393
  31. Ho, P.T.P. et al. 1991, NATURE 350, 309
  32. Jackson, J. et al. 1993, Ap.J. 402, 173
  33. Lo, K.Y. and Claussen, M, J. 1983, NATURE 306, 647
  34. Serabyn, E. and Lacy, J. 1985, Ap.J. 293, 445
  35. Serabyn, E., Lacy, J., Townes, C.H. and Bharat,R. 1988, Ap.J. 326, 171
  36. Geballe, T.R. et al. 1984, Ap.J. 284, 118
  37. Yusef-Zadeh, F. 1994, in (38), 355
  38. Genzel, R. and Harris, A.l. 1994, Nuclei of Normal Galaxies: Lessons from the Galactic Center, (Dordrecht:Kluwer)
  39. Lutz, D., Krabbe, A. and Genzel, R. 1993, Ap.J. 418, 214
  40. Aitken, D.K. et al. 1991, Ap.J. 380, 419
  41. Hildebrand, R. H. et al. 1993, Ap.J. 417, 565
  42. Schwarz, U. and Lasenby, J. 1990, in Galactic end Extragalactic Magnetic Fields, eds. R.Beck, P.Kronberg and R.Wielebinski (Dordrecht:Kluwer), 383
  43. Killeen, N.E.B., Lo, K.Y. and Crutcher,R. 1992, Ap.J. 385, 585
  44. Ekers, R.D., van Gorkom, J.H., Schwarz, U. and Goss, W.M. 1983, Astr.Ap. 122, 143
  45. Mezger, P.G.et al. 1989, Astr.Ap. 209, 337
  46. Zylka,R., Mezger, P.G. and Wink, J. 1990, Astr.Ap. 234, 133
  47. Allen, D.A. 1994, in (38), 293
  48. Rieke, G. and Rieke, M. 1994, in (38), 283
  49. Eckart, A., Genzel, R., Hofmann, R., Sams, B.J. and Tacconi-Garman, L.E. 1993, Ap.J. 407, L77
  50. Phinney, E.S. 1989, in (112), 543
  51. Lebofsky, M.J., Rieke, G.H. and Tokunaga, A. 1982, Ap.J. 263, 736
  52. Allen, D.A., Hyland, A.R. and Hillier, D.J. 1990, MNRAS 244, 706
  53. Krabbe, A., Genzel, R., Drapatz, S. and Rotaciuc, V. 1991, Ap.J. 382, L1 9
  54. Backer, D. 1994, in (38), 403
  55. Davidson, J.A., Werner, M.W., Wu, X., Lester, D.F., Harvey, P.M., Joy, M. and Morris, M. 1992, Ap.J. 387, 189
  56. Becklin, E.E., Gatley, I. and Werner, M.W. 1982, Ap.J. 258, 134
  57. Lacy, J.H., Townes, C.H., Geballe, T.R. and Hollenbach, D. 1980, Ap.J. 241, 132
  58. Shields, J.C. and Ferland, G.J. 1994, Ap.J.430,236
  59. Erickson, E.F., Colgan, W.J., Simpson, J.P., Rubin, R.H. and Haas, M.R. 1994, in (38), 249
  60. Lacy, J.H., Townes, C.H. and Hollenbach, D.J. 1982, Ap.J. 262, 120
  61. Herter, T., Houck, J.R., Shure, M., Gull, G.E. and Graf, P. 1984, Ap.J. 287, L15
  62. Lester, D.F., Dinerstein, H.L., Werner, M.W., Watson, D.M., Genzel, R. and Storey, J.W.V. 1987, Ap.J. 320, 573
  63. Forrest, W.J., Pipher, J.L. and Stein, W.A. 1986, Ap.J. 301, L49
  64. Eckart, A., Genzel, R., Krabbe, A., Hofmann, R., van der Werf, P.P. and Drapatz, S. 1992, NATURE 355, 526
  65. Eckart, A., Genzel, R., Hofmann, R., Sams, B. and Tacconi-Garman, L. 1995, Ap.J. in press
  66. Gerhard, O. 1994, in (38),267
  67. Lee, H.M. 1994, in (38), 335
  68. Forrest, W.J., Shure, M.A., Pipher, J.L. and Woodward, C.A.1987 in "The Galactic Center", ed. D.Backer, AIP Conference Proceedings 155, 153
  69. Krabbe, A. Genzel, R. et al. 1995, in prep.
  70. Blum, R.D., dePoy, D.L. and Sellgren, K. 1995a, Ap.J. in press
  71. Hall, D.N.B., Kleinmann, S.G. and Scoville, N.Z. 1982, Ap.J. 262, L53
  72. Geballe, T.R. et al. 1987, Ap.J. 320, 562
  73. Najarro, F. et al. 1994, Astr.Ap. 285, 573
  74. .
  75. Blum, R.D., Sellgren, K. and DePoy, D.L. 1995b, Ap.J. submitted
  76. Rieke, G.H. and Lebofsky, M.J. 1982, in "The Galactic Center", eds. G.R.Riegler and R.D.Blandford, AIP conference series 83 (New York), 194
  77. Allen, D.A. and Sanders, R.H. 1986, NATURE, 319, 191
  78. Tamblyn, P. and Rieke, G.H. 1993, Ap.J. 414, 573
  79. Morris, M. 1993, Ap.J . 408, 496
  80. Lee, H.M. 1987, Ap.J. 319, 801
  81. Lee, H.M. 1994, in "The Galaxy", ed.L.BIitz (Dordrecht:Kluwer), in press
  82. Meynet, G., Maeder, A., Schaller, G., Schaerer, D. and Charbonnel, C.1994, Astr.Ap.(Suppl.) 103, 97
  83. Langer, N. et al. 1994, Astr.Ap. in press
  84. Haller, J.W. and Rieke, M.J. 1989, in (112), 487
  85. Ekers, R.D. and Lynden-Bell, D. 1971, Ap.Lett. 9, 189
  86. Downes, D. and Martin, A. 1971, NATURE 233, 112
  87. Balick, B. and Brown, R.L. 1974, Ap.J. 194, 265
  88. Lynden-Bell, D. and Rees, M. 1971, MNRAS 152, 461
  89. Lo, K.Y. 1989, in (112),527
  90. Krichbaum, T.P., Schalinski, C.J., Witzel, A., Standke, K.J., Graham, D.A. and Zensus, J.A. 1994, in (38), 411
  91. Zylka, R., Mezger, P.G. and Lesch, J. 1992, Astr.Ap.261, 119
  92. Watson, M.G., Willingale, R., Grindlay, J.E. and Hertz, P. 1981, Ap.J. 250, 142
  93. Predehl, P. and Trumper, J. 1994, Astr.Ap. in press
  94. Skinner, G.K. et al. 1987, NATURE 330, 544
  95. Goldwurm et al., 1994, NATURE 371, 589
  96. Mezger, P.G. 1994, in (38), 41 5
  97. Genzel, R. 1987, in Proceedings of 13th Texas Symposium, ed. M.P.Ulmer (World Scientific, Singapore), 388
  98. Lacy, J.H., Achtermann, J.M. and Serabyn, E. 1991, Ap.J. 380, L71
  99. Herbst, T.M., Beckwith, S.V.W., Forrest, W.J. and Pipher, J.L. 1993, A.J. 105, 956
  100. Sellgren, K., McGinn, M.T., Becklin, E. and Hall, D.N.B. 1990, AP.J. 359, 112
  101. Haller, J.W., Rieke, M.J. and Rieke, G.H. 1992, B.A.A.S. 24, 1178
  102. Krabbe, A. et al. 1994, in "Infrared Astronomy with Arrays (3)", ed. I.McLean (Dordrecht:Kluwer), 484
  103. Kormendy, J. 1994, in (38), 379
  104. Miyoshi, M., Moran, J.M., Hernstein, J., Greenhill, L., Naakai, N., Diamond, P. and Inoue, M. 1995, NATURE 373, 127
  105. Rees, M. 1988, NATURE 333, 523
  106. Melia, F. 1992, Ap.J. 387, L25
  107. Ozernoy, L. and Genzel, R. 1995, in "The Galaxy", ed.L.BIitz (Kluwer: Dordrecht),in press
  108. Lindqvist, M., Habing, H. and Winnberg, A. 1992, Astr.Ap. 259, 118
  109. Rieke, G.H. and Rieke, M.J. 1988, Ap.J. 330, L33
  110. Gusten, R. et al. 1987, Ap.J. 318, 124
  111. Schwarz, U.J., Bregman, J.D. and van Gorkom, J.H. 1989, Astr.Ap.215, 33
  112. Morris, M. (ed.) 1989, "The Center of the Galaxy" (Dordrecht:Kluwer)