Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://hea-www.harvard.edu/~garcia/apj.553.L47.ps Äàòà èçìåíåíèÿ: Wed Feb 9 00:13:34 2011 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 21:18:47 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: http astrokuban.info astrokuban

L47
The Astrophysical Journal, 553:L47--L50, 2001 May 20
# 2001. The American Astronomical Society. All rights reserved. Printed in U.S.A.
NEW EVIDENCE FOR BLACK HOLE EVENT HORIZONS FROM CHANDRA
Michael R. Garcia, 1 Jeffrey E. McClintock, 1 Ramesh Narayan, 1 Paul Callanan, 2
Didier Barret, 3 and Stephen S. Murray 1
Received 2000 December 21; accepted 2001 April 19; published 2001 May 8
ABSTRACT
Previously we claimed that black hole X­ray novae (BHXNs) in quiescence are much less luminous than
equivalent neutron star X­ray novae (NSXNs). This claim was based on the quiescent detection of a single short­
period BHXN (A0620#00, hr) and two longer period BHXNs (GRO J1655#40, hr; V404
P p 7.8 P p 62.9
orb orb
Cygni, hr), along with sensitive upper limits. Here we announce the detection of two more short­
P p 155.3
orb
period BHXNs (GRO J0422#32, hr; GS 2000#25, hr), an upper limit for a third that is
P p 5.1 P p 8.3
orb orb
improved by 2 orders of magnitude (4U 1543#47, hr), and a new, much lower quiescent measurement
P p 27.0
orb
of GRO J1655#40. Taken together, these new Chandra Advanced CCD Imaging Spectrometer measurements
confirm that the quiescent X­ray luminosities of BHXNs are significantly lower than those of NSXNs. We argue
that this provides strong evidence for the existence of event horizons in BHXNs.
Subject headings: binaries: close --- black hole physics ---
stars: individual (A0620#00, GRO J0422#32, GRO J1655#40, GS 2000#25,
V404 Cygni, 4U 1543#47) --- stars: neutron --- X­rays: stars
1. INTRODUCTION
The very low, but nonzero, quiescent X­ray luminosity of
the black hole X­ray nova (BHXN) A0620#00 is difficult to
understand in the context of standard viscous accretion disk
theory (McClintock, Horne, & Remillard 1995), given the con­
tinued mass transfer from the companion evidenced by an op­
tically bright disk. The observations can be explained by an
advection­dominated accretion flow (ADAF) model (Narayan,
McClintock, & Yi 1996). In an ADAF (see Narayan, Mahad­
evan, & Quataert 1998b and Kato, Fukue, & Mineshige 1998
for reviews), the energy released by accretion is stored as heat
in a radiatively inefficient flow. Should the accreting object be
a black hole (BH), this energy would be lost from view once
it crosses the event horizon, but if the object has a solid surface,
the energy would be released upon impact with that surface
and radiated to infinity. Thus, for the same mass accretion rate,
a BH would be significantly less luminous than a compact star
with a surface (Narayan & Yi 1995). Such a comparison of
otherwise similar systems is a promising method for proving
the reality of event horizons (Narayan, Garcia, & McClintock
1997b, hereafter NGM97).
X­ray novae (XNs) containing BH and neutron star (NS)
primaries are believed to be similar in many respects. Most
importantly, the mass transfer rate from the secondary, mea­
sured in Eddington units of the primary, is believed to be
comparable in BHXNs and neutron star X­ray novae (NSXNs)
of similar orbital periods (Menou et al. 1999, hereafter M99).
NGM97, Garcia et al. (1998, hereafter G98), and M99 showed
that quiescent BHXNs are much less luminous in X­rays than
quiescent NSXNs of similar orbital periods and argued that
this provides direct evidence that BHXNs are able to ``hide''
their accretion energy behind an event horizon. However, their
claims were based on a very small sample in which only one
1 Harvard­Smithsonian Center for Astrophysics, 60 Garden Street, Cam­
bridge, MA 02138; mgarcia@cfa.harvard.edu, jmcclintock@cfa.harvard.edu,
rnarayan@cfa.harvard.edu, smurray@cfa.harvard.edu.
2 Department of Physics, University College, College Road, Cork, Ireland;
paulc@ucc.ie.
3 Centre d'Etude Spatiale des Rayonnements, 9 Avenue du Colonel Roche,
31028 Toulouse, France; didier.barret@cesr.fr.
detected BHXN (A0620#00, hr) had an orbital pe­
P p 7.8
orb
riod similar to those of the comparison NSXNs. We report here
new observations of six BHXNs with the Chandra X­Ray Ob­
servatory (Weisskopf & O'Dell 1997).
2. OBSERVATIONS AND ANALYSIS
The data described herein were obtained with the Advanced
CCD Imaging Spectrometer (ACIS; Garmire et al. 1992) on
board Chandra. The Chandra/ACIS combination is #100 times
more sensitive than previous X­ray observatories, making it
ideal for studying the extremely faint quiescent state of BHXNs.
The observations are listed in Table 1. Data were analyzed with
a combination of the Chandra X­Ray Center (CXC) CIAO
v1.1 software, 4 the HEASARC XSPEC v11.0 software, 5 and
software written by A. Vikhlinin (Vikhlinin et al. 1998). De­
tected fluxes and count rates were restricted to a 0.3--7.0 keV
energy band for ACIS­S observations, and 0.5--7.0 keV for
ACIS­I observations, in order to reduce the instrumental back­
ground. The observed on­axis point­source response function
is such that 95% of the source flux is contained in the 1#.5
radius source extraction circle (van Speybroeck et al. 1997 6 ).
We defer a discussion of the X­ray spectra of quiescent
BHXNs, as determined by the ACIS, to a second paper (J. E.
McClintock, M. R. Garcia, R. Narayan, & S. S. Murray 2001,
in preparation). Here we assume that the spectra of all the
sources are adequately described by a single power law with
and absorption consistent with the optical extinction (Na­
a # 2
rayan, Barret, & McClintock 1997a). Our conclusions are un­
changed for any reasonable assumption of the source spectra
(e.g., Narayan et al. 1997a).
GRO J0422#32.---The GRO J0422#32 region was ob­
served with the Chandra ACIS­I CCD array (Garmire et al.
1992) for 18.8 ks on 2000 October 10. The CCD array recorded
4 The CXC Data Manipulation User's Guide, Version 1.1; available at
http://asc.harvard.edu/ciao/download/doc/manual_dm.ps.
5 K. Arnaud & B. Dorman 2000, XSPEC, An X­Ray Spectral Fitting Pack­
age, Users Guide for Version 11; available at http://heasarc.gsfc.nasa.gov/docs/
xanadu/xspec/index.html.
6 See also the Chandra Proposers Observatory Guide, Version 2.0; available
at http://asc.harvard.edu/udocs/docs/docs.html.

L48 CHANDRA OBSERVATIONS OF BLACK HOLES Vol. 553
TABLE 1
Chandra Observations of Quiescent BHXNs (Detected Fluxes and Limits)
System Instrument Counts
Exposure
(ks)
f X
(ergs s #1 cm #2 )
L X
(ergs s #1 )
D
(kpc)
m 1
(M , )
log N H
(cm #2 )
GRO J0422#32 . . . . . . . . . . . . . . . . . . ACIS­I 16 18.8 6.6 # 10 #15 7.6 # 10 30 2.6 12 21.3
A0620#00 (XN Mon 75) . . . . . . . ACIS­S 123 38.3 1.8 # 10 #14 2.8 # 10 30 1.0 6.1 21.3
GS 2000#25 (XN Vul 88) . . . . . . ACIS­I 5 21.5 2.4 # 10 #15 2.4 # 10 30 2.7 8.5 21.9
4U 1543#47 . . . . . . . . . . . . . . . . . . . . . ACIS­I !5 9.9 !4.2 # 10 #15 !3.0 # 10 31 6.1 a 6 a 21.6 b
GRO J1655#40 . . . . . . . . . . . . . . . . . . ACIS­S 65 42.5 1.2 # 10 #14 2.4 # 10 31 3.2 7 21.8
V404 Cyg (GS 2023#338) . . . . . . ACIS­S 1655 10.3 1.5 # 10 #12 4.9 # 10 33 3.5 12 22.0
Note.---Values of N H are taken from G98 and NGM97; distances and primary masses are from M99, unless explicitly mentioned.
Detected ACIS­S fluxes are over a 0.3--7.0 keV band, ACIS­I fluxes are over a 0.5--7.0 keV band, and emitted luminosities are over
a 0.5--10.0 keV band.
a From J. Orosz et al. 2001, in preparation.
b N H from Greiner, Predehl, & Harmon 1994 and van der Woerd, White, & Kahn 1989.
16 photons at a position consistent with the optical position of
GRO J0422#32. Given the extremely low background of
Chandra (!1 background count per target for every observation
presented here), the recording of #5 counts represents a de­
tection of the target with greater than 99% confidence. Assum­
ing a distance of 2.6 kpc (M99) and X­ray absorption corre­
sponding to the optical extinction of (Filippenko,
A p 1.2
V
Matheson, & Ho 1995; Predehl & Schmitt 1995), we derive
an emitted luminosity of ergs s #1 (0.5--10.0 keV).
30
7.6 # 10
A0620#00.---The A0620#00 region was observed for 45 ks
on 2000 February 29. We rejected 6.7 ks of this observation
because of intervals of enhanced background (see Plucinsky
& Virani 2000). A0620#00 was placed on the ACIS­S3 CCD
(Garmire et al. 1992) in order to maximize the count rate from
the possibly soft X­ray spectrum (Narayan et al. 1997a). The
CCD recorded 123 photons at a position consistent with the
optical position of A0620#00.
GS 2000#25.---The GS 2000#25 region was observed with
the Chandra ACIS­I detector on 1999 November 5 for a total
of 21.5 ks. The CCD array recorded 5 photons near the position
of GS 2000#25, which is a weak but significant detection. In
order to more accurately determine the position of this source,
we cross­correlated the positions of all sources in the image
with greater than 10 counts against an optical astrometric ref­
erence catalog (USNO­A2; Monet et al. 1996). We found four
matches within a 2# search radius. We then determined the
astrometric solution for the X­ray image using these four stars.
This solution has an rms position uncertainty of 0#.3 and shows
that the position of this weak source is ,
h m s
R.A. p 20 02 49.52
(J2000). Our ability to determine the
#
decl. p #25#14 10#.34
centroid of this source is limited by the low number of counts
to #0#.5. We determined the position of the optical counterpart
of GS 2000#25 in a similar way, matching eight USNO­A2
stars against a K­band image of the field we acquired at the
Lick 3 m telescope in California with the Gemini camera. This
plate solution has an rms position uncertainty of 0#.2 and shows
the position of the optical (IR) counterpart to be R.A. p
, (J2000). The offset be­
h m s #
20 02 49.55 decl. p #25#14 10#.94
tween the two positions is ; thus, the position of the
0#.7 # 0#.8
weak X­ray source is consistent with GS 2000#25.
4U 1543#47.---A 9.9 ks ACIS­I observation of 4U 1543#47
took place on 2000 July 26 and failed to detect even a single
photon from this source. We compute a conservative upper
limit to the luminosity by assuming that less than 5 photons
were detected. This luminosity (see Table 1) is a factor of #100
below the previous upper limit of ergs s #1 (M99).
33
2 # 10
GRO J1655#40.---A 42.5 ks ACIS­S observation of GRO
J1655#40 took place on 2000 July 2 and detected 65 photons
from this source. We note that this corresponds to a luminosity
a factor of #10 below the previously measured quiescent lu­
minosity of ergs s #1 (G98). Given the quiescent
32
2.5 # 10
variations seen in V404 Cygni (Wagner et al. 1994), this could
simply be indicative of typical quiescent variations. Alternately,
we note that the previous quiescent observations were in be­
tween two large outbursts separated by #1 yr and therefore
may not have been indicative of the true quiescent level. The
Chandra observations reported here occurred #4 yr after the
last outburst, and therefore may measure the quiescent level
more accurately.
V404 Cyg.---A 10.3 ks ACIS­S observation of V404 Cyg
took place on 2000 April 26 and allowed detection of 1655
photons from this source.
3. DISCUSSION
We have collected in Table 2 the available quiescent X­ray
luminosities and upper limits for BHXNs and NSXNs. We have
supplemented our Chandra measurements with the recent qui­
escent detections of SAX J1808.4#3658 (Stella et al. 2000;
Dotani, Asai, & Wijnands 2000) and included previous data
compiled by M99. Figure 1 displays the Eddington­scaled lu­
minosities as a function of orbital period P orb . For calculating
L Edd , we used the BH mass estimates in Table 1 and assumed
that all NSs have a mass of .
1.4 M,
With the Chandra detections of GRO J0422#32 and GS
2000#25 reported here, three short­period BHXNs have now
been detected in quiescence. A greatly improved upper limit
for a fourth short­period system, 4U 1543#47, is nearly com­
parable to these detections. In addition, we find a new, 10 times
fainter quiescent luminosity for GRO J1655#40, which has a
period intermediate between the NSXNs and the long­period
BHXN V404 Cyg.
The new data points significantly strengthen our earlier claim
(NGM97; G98; M99) that BHXNs have much lower (roughly
a factor of 100) quiescent luminosities than NSXNs. As ex­
plained in § 1, such a difference in luminosity is natural if
quiescent accretion proceeds via a radiatively inefficient ADAF
and if the primaries in BHXNs have event horizons. Thus, the
new data bolster the evidence for event horizons in BHXNs.
Our argument relies on the reasonable assumption that the
Eddington­scaled mass accretion rate is similar in quiescent
BHXNs and NSXNs. At the short orbital periods characteristic
of the NSXNs in our sample, angular momentum loss through
gravitational radiation is expected to be the dominant mecha­
nism driving mass transfer from the secondary. The Eddington­
scaled mass transfer rates are then likely to be roughly similar

No. 1, 2001 GARCIA ET AL. L49
TABLE 2
Quiescent Luminosities
System
(1)
P orb
(hr)
(2)
log L min
(ergs s #1 )
(3)
Neutron Star Primaries
SAX J1808#365 . . . . . . . . . . 2.0 a 32.0, b 31.5 c
EXO 0748#676 . . . . . . . . . . . 3.82 34.1
4U 2129#47 . . . . . . . . . . . . . . . 5.2 32.8
Cen X­4 . . . . . . . . . . . . . . . . . . . . 15.1 32.4
Aql X­1 . . . . . . . . . . . . . . . . . . . . 19 32.6
H1608#52 . . . . . . . . . . . . . . . . . 12 d 33.3
Black Hole Primaries
GRO J0422#32 . . . . . . . . . . . 5.1 30.9, !31.6
A0620#00 . . . . . . . . . . . . . . . . . 7.8 30.5, 30.8 e
GS 2000#25 . . . . . . . . . . . . . . 8.3 30.4, !32.2
GS 1124#683 . . . . . . . . . . . . . 10.4 !32.4
H1705#250 . . . . . . . . . . . . . . . . 12.5 !33.0
4U 1543#47 . . . . . . . . . . . . . . . 27.0 !31.5, !33.3
GRO J1655#40 . . . . . . . . . . . 62.9 31.3, 32.4
V404 Cyg . . . . . . . . . . . . . . . . . . 155.3 33.7, 33.1
Note.---References are as in M99, unless otherwise noted. New
and/or Chandra measurements are in boldface. Col. (2): Orbital
period. Col. (3): Luminosity in quiescence in the 0.5--10 keV band
(corrected for the revised distances).
a From Chakrabarty & Morgan 1998.
b From Stella et al. 2000; assuming kpc.
D p 2.5
c From Dotani et al. 2000.
d Wachter 2000 reports a new 12 hr period, intermediate between
the previously reported 98.4 hr (Ritter & Kolb 1998) and 5 hr (Chen
et al. 1998) periods.
e Recomputed for from NGM97.
a p 2
f From Kong 2000.
Fig. 1.---Top: Quiescent luminosities of BHXNs ( filled circles) and NSXNs
(open circles) from Table 2. Data points not included in M99 are circled.
Multiple quiescent detections are included. Bottom: Only the lowest quiescent
detections or Chandra upper limits.
in BHXNs and NSXNs of similar P orb (M99). At the long
periods of V404 Cyg and perhaps GRO J1655#40, nuclear
evolution is expected to drive the mass transfer rate to sub­
stantially higher values, and so these systems are less useful
for our purposes. Moreover, the lack of NSXNs with similar
P orb limits the usefulness of these systems for the comparisons
made here.
We also assume that , the fraction of matter transferred
f #
from the secondary that actually reaches the central star, is the
same in BHXNs and NSXNs. Outflowing winds from an ADAF
(Narayan &Yi 1994, 1995; Blandford &Begelman 1999; M99)
tend to reduce , but there is no reason to expect winds to be
f #
stronger (by a factor of 100) in BHXNs than in NSXNs. In
fact, a centrifugal propeller and/or radio pulsar action could
reduce in NSXNs without affecting BHXNs (NGM97; M99).
f #
Therefore, we expect to be lower in NSXNs than in BHXNs
f #
rather than the other way round. This strengthens the case for
event horizons in BHXNs.
In this connection, we note that SAX J1808.4#3658 is the
least luminous of the NSXNs. It also may have a time­averaged
mass transfer rate that is lower than that typical of XNs, perhaps
due to the action of the radio pulsar on the secondary (Chak­
rabarty & Morgan 1998) or due to the possible extreme evo­
lution of the secondary (King 2000; Dotani et al. 2000). Given
that SAX J1808.4#3658 is the only example of an NSXN that
shows permanent coherent pulsations, it seems possible that it
may also have an unusually efficient ``propeller.'' Despite all
these anomalies, this NSXN is #60 times brighter than our
three short­period BHXNs (see Fig. 1).
Campana & Stella (2000) argue that it is not correct to com­
pare only X­ray luminosities, as we have done, but that we
should include also the quiescent nonstellar optical and UV
luminosity. This point is not obvious since the origin of the
optical/UV luminosity is presently unclear. Within the ADAF
model, it depends on the poorly known radius at which the
inner edge of the accretion disk evaporates into the ADAF
(e.g., Narayan et al. 1996, 1997a). It also depends on the
strength of winds from the ADAF (Quataert & Narayan 1999).
If the nonstellar optical/UV luminosity originates in the outer
accretion disk, or in the hot spot where the mass transfer stream
from the secondary impacts the disk, then the fact that the
optical/UV luminosities of BHXNs and NSXNs are similar
(Campana & Stella 2000) provides observational confirmation

L50 CHANDRA OBSERVATIONS OF BLACK HOLES Vol. 553
that the mass transfer rates in the two kinds of system are
similar, as we have assumed. A 0.25 mag modulation in the
far­UV flux of A0620#00 on an orbital timescale suggests that
the far­UV flux may be modulated with the orbital phase
(McClintock 2000), which indicates an origin in the hot spot.
Bildsten & Rutledge (2000; but see Lasota 2000) suggest
that the X­rays in many BHXNs may be produced by a rota­
tionally enhanced stellar corona in the secondary. The detection
of GRO J0422#32 with a luminosity of nearly 10 31 ergs s #1
rules out the coronal interpretation for this system but is con­
sistent with the prediction of the ADAF plus disk instability
model of Lasota (2000). Similarly, the luminosity of V404 Cyg
is #60 times larger than the coronal model predicts but within
a factor of 10 of the ADAF model (Lasota 2000). The lumi­
nosities of the remaining four sources in Table 1 are consistent
with both models. If some (or all) of the X­ray emission in
BHXNs were coronal, then the accretion luminosities of the
BHs would be even lower than our estimates and the argument
for event horizons would be further strengthened.
A critical element in our comparison of NSXN and BHXN
quiescent luminosities is the assumption that the quiescent X­
ray luminosities of NSXNs result from accretion. Brown, Bild­
sten, & Rutledge (1998) have suggested that the luminosity
could be due to crustal heating of the NS during outburst fol­
lowed by cooling in quiescence. The rapid variability of the
prototypical NSXN Cen X­4 (Campana et al. 1997) is hard to
explain in a cooling model and shows that at most approxi­
mately one­third of the quiescent luminosity of this source is
due to crustal cooling (M99). In addition, Cen X­4 and Aql
X­1 have substantial power­law tails in their spectra (carrying
about half the total luminosity), and it is hard to explain this
spectral component with a cooling model. It thus seems rea­
sonable to assume that accretion accounts for a substantial frac­
tion of the quiescent X­ray luminosity in most NSXNs. The
optical variability of NSXNs in quiescence (McClintock &
Remillard 2000; Jain et al. 2000; Ilovaisky & Chevalier 2000)
provides ample evidence that accretion continues during
quiescence.
Regardless of the caveats mentioned above, ultimately the
dramatic difference in quiescent X­ray luminosities of BHXNs
and NSXNs needs to be explained. Any explanation is likely
to require a fundamental difference in the nature of the pri­
maries in BHXNs and NSXNs. In our view, any straightforward
explanation, whether based on ADAFs or not, will require pos­
tulating an event horizon in BHXNs.
The Eddington­scaled luminosities of some supermassive
black holes (SMBHs) in galactic nuclei (e.g., Garcia et al. 2000;
Baganoff et al. 2000) are much smaller than those of the ``stellar
mass'' BHs discussed here. Could this be evidence for the event
horizon (Narayan et al. 1998a)? Unfortunately, the mass ac­
cretion rates of SMBHs are difficult to determine observation­
ally (DiMatteo et al. 2001b; DiMatteo, Carilli, & Fabian 2001a;
Quataert, Narayan, & Reid 1999), and one cannot exclude the
possibility that the low luminosities of SMBHs are simply the
result of extremely low accretion rates. In the case of quiescent
BHXNs, we have the luxury of a control sample of NSXNs.
By comparing quiescent BHXNs and NSXNs with similar or­
bital periods, we eliminate the uncertainty in the mass accretion
rate and thus obtain more secure evidence for the event horizon.
This work was supported in part by NASA contract NAS8­
39073 to the Chandra X­Ray Center, contract NAS8­38248 to
the HRC Team, and NSF grant AST 98­20686 to R. N. We
thank A. Kong for helpful comments.
REFERENCES
Baganoff, F., et al. 2000, AAS HEAD Meeting, 32, 02.10
Bildsten, L., & Rutledge, R. E. 2000, ApJ, 541, 908
Blandford, R. D., & Begelman, M. C. 1999, MNRAS, 303, L1
Brown, E. F., Bildsten, L., & Rutledge, R. E. 1998, ApJ, 504, L95
Campana, S., Mereghetti, S., Stella, L., & Colpi, M. 1997, A&A, 324, 941
Campana, S., & Stella, L. 2000, ApJ, 541, 849
Chakrabarty, D., & Morgan, E. H. 1998, Nature, 394, 346
Chen, W., et al. 1998, in AIP Conf. Proc. 431, Accretion Processes in Astro­
physical Systems: Some Like It Hot!, ed. S. S. Holt & T. R. Kallman
(Woodbury: AIP), 347
DiMatteo, T., Carilli, C. L., & Fabian, A. C. 2001a, ApJ, 547, 731
DiMatteo, T., Johnstone, R. M., Allen, S. W., & Fabian, A. C. 2001b, ApJ,
550, L19
Dotani, T., Asai, K., & Wijnands, R. 2000, ApJ, 543, L145
Filippenko, A. V., Matheson, T., & Ho, L. C. 1995, ApJ, 455, 614
Garcia, M. R., McClintock, J. E., Narayan, R., & Callanan, J. 1998, in ASP
Conf. Ser. 137, Wild Stars in the Old West, ed. S. Howell, E. Kuulkers, &
C. Woodward (San Francisco: ASP), 506 (G98)
Garcia, M. R., Murray, S. S., Primini, F. A., Forman, W., & Jones, C. 2000,
ApJ, 537, L23
Garmire, G. P., et al. 1992, in AIAA Space Programs and Technologies Conf.
(Reston: AIAA), 8
Greiner, J., Predehl, P., & Harmon, B. A. 1994, in AIP Conf. Proc. 304, Proc.
Second Compton Symp., ed. C. E. Fichtel, N. Gehrels, & J. P. Norris (New
York: AIP), 314
Ilovaisky, S. A., & Chevalier, C. 2000, IAU Circ. 6025
Jain, R., et al. 2000, IAU Circ. 7429
Kato, S., Fukue, J., &Mineshige, S. 1998, Black Hole Accretion Disks (Kyoto:
Kyoto Univ. Press)
King, A. R. 2000, MNRAS, 315, L33
Kong, A. 2000, Ph.D. thesis, Oxford Univ.
Lasota, J. P. 2000, A&A, 360, 575
McClintock, J. E. 2000, in Rossi2000: Astrophysics with the Rossi X­Ray
Timing Explorer, ed. T. E. Strohmayer (Greenbelt: NASA/GSFC), 89
McClintock, J. E., Horne, K., & Remillard, R. A. 1995, ApJ, 442, 358
McClintock, J. E., & Remillard, R. R. 2000, ApJ, 531, 956
Menou, K., Esin, A. A., Narayan, R., Garcia, M. R., Lasota, J. P., & Mc­
Clintock, J. E. 1999, ApJ, 520, 276 (M99)
Monet, D., et al. 1996, The USNO­A2.0 Catalog (Washington, DC: USNO)
Narayan, R., Barret, D., & McClintock, J. E. 1997a, ApJ, 482, 448
Narayan, R., Garcia, M. R., & McClintock, J. E. 1997b, ApJ, 478, L79
(NGM97)
Narayan, R. Mahadevan, R., Grindlay, J. E., Popham, R. G., & Gammie, C.
1998a, ApJ, 492, 554
Narayan, R., Mahadevan, R., & Quataert, E. 1998b, in Theory of Black Hole
Accrection Disks, ed. M. A. Abramowicz, G. Bjornsson, & J. E. Pringle
(Cambridge: Cambridge Univ. Press), 148
Narayan, R., McClintock, J. E., & Yi, I. 1996, ApJ, 457, 821
Narayan, R., & Yi, I. 1994, ApJ, 428, L13
---------. 1995, ApJ, 444, 231
Plucinsky, P. P., & Virani, S. N. 2000, Proc. SPIE, 4012, 669
Predehl, P., & Schmitt, J. H. M. M. 1995, A&A, 293, 889
Quataert, E., & Narayan, R. 1999, ApJ, 520, 298
Quataert, E., Narayan, R., & Reid, M. J. 1999, ApJ, 517, L101
Ritter, H., & Kolb, U. 1998, A&AS, 129, 83
Stella, L., Campana, S., Mereghetti, S., Ricci, D., & Israel, G. L. 2000, ApJ,
537, L115
van der Woerd, H., White, N. E., & Kahn, S. M. 1989, ApJ, 344, 320
van Speybroeck, L., Jerius, D., Edgar, R. J., Gaetz, T. J., Zhao, P., & Reid,
P. B. 1997, Proc. SPIE, 3113, 89
Vikhlinin, A., McNamara, B. R., Forman, W., Jones, C., Quintana, H., &
Hornstrup, A. 1998, ApJ, 502, 558
Wachter, S. 2000, in 1999 Pacific Rim Conference on Stellar Astrophysics,
ed. K. S. Cheng (Dordrecht: Kluwer), 451
Wagner, M. R., Starrfield, S. G., Hjellming, R. M., Howell, S. B., & Kreidl,
T. J. 1994, ApJ, 429, L25
Weisskopf, M. C., & O'Dell, S. L. 1997, Proc. SPIE, 3113, 2