0807.1547/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: 
3: \shorttitle{Observational Comparison between ULXs and XRBs}
4: \shortauthors{Berghea et al.}
5: 
6: \begin{document}
7: 
8: 
9: \title{Testing the Paradigm that Ultraluminous X-ray Sources as a Class 
10: Represent Accreting Intermediate-Mass Black Holes}
11: 
12: 
13: \author{C. T. Berghea}
14: \affil{Department of Physics, The Catholic University of America, Washington, DC 20064}
15: \email{79berghea@cua.edu}
16: 
17: \author{K. A. Weaver}
18: \affil{Laboratory for High Energy Astrophysics, NASA's Goddard Space Flight Center, Greenbelt, MD 20771}
19: \email{kweaver@milkyway.gsfc.nasa.gov}
20: 
21: \author{E. J. M. Colbert}
22: \affil{Department of Physics and Astronomy, Johns Hopkins University, Baltimore, MD 21218}
23: \email{colbert@jhu.edu}
24: \and
25: 
26: \author{T. P. Roberts}
27: \affil{Department of Physics, Durham University, South Road, Durham DH1 3LE, UK}
28: \email{t.p.roberts@durham.ac.uk}
29: 
30: \begin{abstract}
31: 
32: To test the idea that ultraluminous X-ray sources (ULXs) in external 
33: galaxies represent a class of accreting intermediate-mass black holes 
34: (IMBHs), we have undertaken a program to identify ULXs and a lower 
35: luminosity X-ray comparison sample with the highest quality data in the 
36: {\it Chandra} archive. We establish as a general property of ULXs that the 
37: most X-ray-luminous objects possess the flattest X-ray spectra (in the {\it Chandra} bandpass). 
38: No prior sample studies have established the general hardening of ULX spectra 
39: with luminosity. This hardening occurs at the highest luminosities 
40: (absorbed luminosity $\geq$5$\times$10$^{39}$~erg~s$^{-1}$)
41: and is in line with recent models arguing that ULXs are 
42: actually stellar-mass black holes. From spectral modeling, we show that 
43: the evidence originally taken to mean that ULXs are IMBHs - i.e., the 
44: ``simple IMBH model'' - is nowhere near as compelling when a large sample 
45: of ULXs is looked at properly. 
46: During the last couple of years, {\it XMM-Newton}
47: spectroscopy of ULXs has to a large extent begun to negate the simple 
48: IMBH model based on fewer objects. We confirm and expand these results, 
49: which validates the {\it XMM-Newton} work in a broader sense with independent X-ray 
50: data. We find that (1) cool-disk components are present with roughly 
51: equal probability and total flux fraction for any given ULX, regardless 
52: of luminosity, and (2) cool-disk components extend below the 
53: standard ULX luminosity cutoff of 10$^{39}$~erg~s$^{-1}$, 
54: down to our sample limit of 10$^{38.3}$~erg~s$^{-1}$.
55: The fact that cool disk components are not correlated with luminosity 
56: damages the argument that cool disks indicate IMBHs in ULXs, for which 
57: strong statistical support was never found. 
58: 
59: \end{abstract}
60: 
61: \keywords{galaxies: general --- surveys --- X-rays:binaries ---
62: accretion, accretion discs}
63: 
64: 
65: \section{INTRODUCTION}
66: 
67: Ultraluminous X-ray sources (ULXs) have long been hailed as direct
68: observational evidence for the existence of accreting
69: intermediate-mass black holes \citep[IMBHs;][]{col99}. The X-ray
70: spectral model that has emerged as a central pillar for this argument
71: (the ``simple IMBH model'') is that which is commonly applied
72: as the canonical X-ray spectral fit to Galactic black-hole binaries
73: with stellar mass black holes \citep{mcc06}.  This
74: model consists of a thermal accretion disk component plus a power-law (PL)
75: continuum component. When applied to ULX spectra, the derived disk
76: temperatures are 0.1$-$0.3~keV \citep[e.g.][]{mill04}, much lower
77: than Galactic black holes (at 0.6$-$1~keV). A cooler disk implies a
78: bigger disk; so assuming that the disk approximately extends inward to the
79: last stable orbit around the black hole, this would imply bigger, and
80: more massive black holes. Such cool disks were indeed found in a few ULXs \citep[e.g.][]{mill04}.
81: 
82: To counter this argument, many recent papers have pointed out both
83: theoretical and observational problems with the simple IMBH model as a
84: global explanation for all ULXs \citep[e.g.][]{gon06, wil06, rob07}.
85: The observed accretion disk components can be fairly
86: weak; thus they do not provide a reliable measure of black hole
87: mass. Also, the simple IMBH model does not necessarily approximate
88: well the X-ray spectra of many ULXs. Attention has switched to perhaps
89: less exotic models to explain some of the ULXs, such as beaming
90: \citep{king01} or super-Eddington accretion \citep{beg02}, both of
91: which explain ULX X-ray properties without the need for an
92: IMBH. Galactic super-Eddington sources are known, such as stellar-mass
93: black hole binaries like GRS~1915+105 \citep{fen04}, V4641~Sgr
94: \citep{rev02} and possibly SS~433. The latter could be an example of
95: both beaming {\it and} super-Eddington emission, the combination of
96: which could easily explain even the most luminous ULXs \citep{beg06, pou07}.
97: Cool accretion disks can also be physically explained by ``coupled disk-corona'' models
98: \citet{don06}, blurred emission and absorption lines from surrounding
99: (outflowing) gas \citep{gon06}, or a microblazar with magnetized jets \citep{fre06},
100: that can transfer disk energy into the jet (thus making the disk
101: fainter and cooling it the same time).
102: 
103: Recent detailed X-ray spectral modeling has revealed properties that
104: further complicate any simple global interpretation, suggesting that
105: multiple classes of ULXs exist.  Some very bright ULXs have been found
106: by several authors \citep{zez02, sor07, sorw06, soc06} to have relatively
107: flat spectra, not usually expected in high states for accretion states
108: of black holes \citep{mcc06}. A flat spectrum suggests an inverse
109: correlation between the slope of the spectrum and source luminosity
110: \citep[see also NGC~5204 X-1;][]{rob06}. Such an inverse correlation
111: is hard to explain with current IMBH models, because in the typical
112: high state the spectrum is soft, dominated by the disk component and
113: with a steep PL \citep{mcc06}. Specifically, {\it XMM-Newton} spectroscopy
114: of ULXs has to a large extent already begun to directly negate the
115: simple IMBH model \citep[see, e.g.][]{wil06, gon06}.
116: 
117: Two {\it Chandra} surveys suggest that ULXs may in fact be an extension 
118: of normal lower-luminosity galaxy populations to higher luminosities. 
119: Using simple PL models applied to spectra with typically ~50 
120: counts each, \citet{swa04} compared ULXs to a lower luminosity sample of 
121: X-ray sources with L$_{X}=$~10$^{38}-$10$^{39}$~erg~s$^{-1}$ and found 
122: both distributions of photon indices to be well fitted by Gaussians 
123: centered at about 1.9. The samples also have similar X-ray colors, time 
124: series and positions within their host galaxies. In another {\it Chandra} study, 
125: \citet{col04} found no discernible difference between the 
126: X-ray colors of ULXs and lower luminosity sources in spiral galaxies. 
127: Both analyses were done with data of fairly poor spectral quality in 
128: terms of fitting detailed models and the latter relied on a color-color 
129: analysis rather than spectral fitting. These works also did not 
130: include two-component model fits that would identify spectral states 
131: and directly test the simple IMBH model
132: 
133: To search carefully for spectral properties that can differentiate
134: ULXs, we need the best data available that will allow us to
135: distinguish between simple spectral models. In this paper we use the
136: highest quality X-ray spectra for a large, complete sample of ULXs
137: from the {\it Chandra} archives to test various ULX models.  We are able to
138: provide a statistically strong comparison of the results with
139: lower luminosity X-ray sources of equal data quality.  We pay special
140: attention to properties usually associated with ULXs, such as the
141: signature of a cool disk, which has never been tested for a uniform
142: and large sample of good quality {\it Chandra} spectra. Is the cool
143: disk preferentially found in ULXs? If yes, does the disk dominate the
144: total emission? We also search for other spectral behaviors found more
145: recently in individual ULXs, such as a correlation between hardness
146: and luminosity, and what this might mean.  For the first time, strong
147: statistical tests of various ideas of ULX models can be provided to
148: the ULX community.
149: 
150: With its unmatched spatial resolution, {\it Chandra} is better suited
151: than {\it XMM-Newton} for studying point sources in crowded regions
152: or resolving point sources in distant galaxies.  This is particularly
153: true for the starburst galaxies that host populations of ULXs (e.g NGC
154: 3256, Lira et al. 2002; Cartwheel galaxy, Gao et al. 2003), where only
155: {\it Chandra}'s unparalleled X-ray optics can spatially and spectrally
156: resolve the emission of ULXs from that of the underlying galaxy. We
157: have searched all public data available in the {\it Chandra} archive
158: for ULXs and lower-luminosity comparison objects with at least 1000
159: counts. In section 2 we present our source selection process, methods
160: for identifying rejected objects, and an estimate of contamination
161: from background objects. In section 3 we discuss the spectral fitting
162: procedures and compare the spectral properties of the two samples. Our
163: goal is to determine whether ULXs as a class have different spectral
164: properties than the less luminous, ``normal'' X-ray sources and to
165: offer an improved diagnosis by using the high-quality spectral data
166: available in the {\it Chandra} archive. In section 4 we present
167: results from the variability analysis. Finally, in Section 5
168: we interpret our results and discuss the insight provided into the
169: nature of ULXs.
170: 
171: 
172: \section{SAMPLE SELECTION AND OBSERVATIONS}
173: 
174: \subsection{Sample Criteria}
175: 
176: There are many published papers that address the nature of ULXs. These 
177: analyses are typically drawn from heterogeneously selected samples, small
178: numbers of objects, or large samples with limited data quality 
179: \citep{swa04, col04}. Comparisons of ULXs with 
180: other types of X-ray point sources in nearby galaxies often use 
181: selection criteria that do not provide the spectral data quality  
182: that allows a robust set of statistical conclusions 
183: to be drawn from data modeling. In this work, we use
184: criteria that create the best possible available sample to address 
185: the nature of ULXs by defining a large and statistically robust sample of
186: ULXs and other pointlike X-ray sources in nearby galaxies with uniform data quality. 
187: Uniformity of data quality is our prime objective, and the completeness of
188: our sample is limited by the observations that are available in the 
189: {\it Chandra} archive, most of which have been obtained by other researchers for a variety of purposes.
190: 
191: Our ULX and comparison samples are statistically robust in the sense that we include
192: all point sources in the {\it Chandra} archive with at least 1000 counts and a
193: luminosity above 10$^{38.3}$~erg~s$^{-1}$. We carefully reject sources 
194: associated with active galactic nuclei, supernovae, and foreground stars. We 
195: also reject piled-up observations to simplify our spectral analysis. 
196: {\it Chandra} provides the most accurate X-ray positions to date; thus
197: we can be sure to identify well-isolated objects for
198: our study. Several {\it XMM-Newton} studies of ULXs are published, but while 
199: these individual spectra are of higher quality, there are fewer individual point sources 
200: available due to {\it XMM-Newton's} poorer imaging resolution and source confusion 
201: for faint targets located in crowded regions in galaxies. We have not 
202: made our sample fully representative in the sense of picking the same
203: number of ULXs and comparison objects from similar galaxy types. Conclusions 
204: about the distribution of objects according to galaxy type can, however, be 
205: inferred from our statistical comparisons. Uniformly good-quality 
206: X-ray spectra allow us to apply exactly the same physical models to the 
207: ULX and comparison samples and directly compare results
208: within the sensitivity limits.
209: 
210: There are selection biases inherent in our analysis. One is distance. 
211: For sources that are intrinsically less luminous, a larger fraction of objects 
212: will be located in the nearest host galaxies,
213: while more luminous objects can be utilized from galaxies at greater
214: distances. We also do not select objects according to any specific 
215: requirement of their local environments (e.g., their locations in their host galaxies). 
216: 
217: 
218: \subsection{Initial Sample}
219: 
220: Our sample is derived from the list of X-ray point sources generated by
221: the XASSIST\footnote{XASSIST (Ptak \& Griffiths 2003)
222: is a semiautomatic X-ray analysis program written and maintained by
223: A.~Ptak. Analysis of archival data processed by XASSIST can
224: be found at http://www.xassist.org} {\it Chandra} pipeline. 
225: For manageability, we have chosen all {\it Chandra} ACIS sources in the public 
226: archives as of a cutoff date of October 18, 2004. We determine 
227: which XASSIST sources are associated with host galaxies following the 
228: procedure used by \citet{col02}. X-ray sources are further considered 
229: if they are located inside the D$_{25}$ ellipse of their
230: host galaxy. Parameters for the D$_{25}$ ellipse 
231: are obtained from v3.9b of the Third Reference Catalog of Bright Galaxies 
232: \citep[RC3;][]{dev91}. We further only consider RC3 galaxies with
233: recessional velocities $cz \le$ 5000 km~s$^{-1}$. 
234: 
235: To estimate observed X-ray luminosities L$_{XA}$,
236: we calculated the 0.3$-$8.0 keV fluxes with XASSIST assuming a PL 
237: model with $\Gamma=$~1.8 and Galactic absorption and used distances 
238: for the associated RC3 galaxies. For galaxies with $cz <$ 1000 km~s$^{-1}$, 
239: distances were taken from \citet{tul88}, otherwise
240: distances were computed using H$_0$~$=$~75~km~s$^{-1}$~Mpc$^{-1}$.
241: We retained all sources with L$_{XA}~>$~10$^{38.3}$~erg~s$^{-1}$
242: and then manually inspected the X-ray images 
243: to eliminate false X-ray sources chosen by the automatic data 
244: processing. This initial selection yielded 126 unique X-ray point 
245: sources in 188 {\it Chandra} observations. A significant fraction 
246: of the point sources were observed multiple times, which provides
247: some useful variability information.
248: 
249: 
250: \subsection{Obvious Rejected objects (AGNs, QSOs, SNe, Stars \& Jets)}
251: 
252: To reject X-ray point sources unrelated to our science, 
253: we used the NASA/IPAC Extragalactic Database (NED)\footnote{Available at http://nedwww.ipac.caltech.edu}. 
254: The absolute positional uncertainty for {\it Chandra} ACIS images is better than 
255: 1$\arcsec$ \citep[e.g.][]{wei03}, which provides the accuracy required 
256: to identify an optical, infrared, or radio counterpart. 
257: Optical positions provided by NED are typically accurate to 
258: within a few arc-seconds, and positions may be slightly less accurate 
259: for infrared and radio sources, so we first searched NED using 
260: a radius of 5$\arcsec$ 
261: surrounding the XASSIST position. We next visually inspected the 
262: X-ray sources and their possible NED counterparts by overlaying the 
263: XASSIST position, the NED position, and the D$_{25}$ galaxy ellipse 
264: onto the raw X-ray images and DSS2\footnote{The Second Digitized 
265: Sky Survey consists of high-resolution scans of several plate collections 
266: in the red, blue, visible, and near-infrared. 
267: The images were downloaded from the server installed at ESO, using a remote 
268: client, the ESO/ST-ECF Digitized Sky Survey application.} 
269: red images. 
270: 
271: Optical images were used to check for bright, foreground 
272: star counterparts. We then refined our identification search by 
273: examining the literature for more accurate positions 
274: for identified NED counterparts. In some cases 
275: VLBI measurements are available with sub-milliarcseconds positional 
276: accuracy, such as those used by \citet{ma98} for the International 
277: Celestial Reference Frame (ICRF). Some published 2MASS positions also 
278: use the ICRF reference system and have accuracies 
279: better than 0.1$\arcsec$, varying slightly with the source brightness 
280: \citep[see UCAC2;][]{zac04}.
281: Optical positions that can be correlated with 
282: radio measurements show systematic differences of only 0.1$\arcsec$ 
283: \citep[e.g.][]{arg90}. Overall, we determined that the 
284: positional uncertainties of identified counterparts are generally much 
285: smaller than our X-ray positional uncertainties, the largest 
286: uncertainty being 1$\arcsec$. We estimated a conservative upper 
287: limit of 1.5$\arcsec$ for the net uncertainty in separation 
288: between the {\it Chandra} X-ray source and an identified optical, IR, or radio 
289: counterpart for any object in our sample. Therefore, we feel 
290: confident that we have identified correct 
291: counterparts to within the errors provided by the X-ray data.
292: 
293: From this search we reject 32 X-ray sources out of 69 {\it Chandra} observations.
294: Most are associated with Seyfert and LINER galaxies \citep{ho97, ver03, bry99}.
295: Others include background quasars and pointlike X-ray 
296: knots associated with jets within the host galaxy. As an example, 
297: source 37 in \citet{zez02}, in the Antennae galaxy pair (NGC~4038/4039), 
298: is a background quasar with redshift 0.26 \citep{cla05}.
299: We identified a supernova in NGC~891 \citep[SN~1986J;][]{bie02}. 
300: One ULX in M101 \citep[NGC 5457 X-6;][]{rob2000} is actually 
301: a foreground star, GSC 2.2~3842. After rejecting sources based
302: on optical and NED counterparts we are left with 
303: 94 X-ray sources in 119 {\it Chandra} observations.
304: 
305: 
306: \subsection{Reprocessing of Archival Data and Final Rejection Criteria}
307: 
308: Having narrowed our sample according to the above criteria, the 
309: ACIS imaging data were retrieved from the {\it Chandra} archives.  
310: The level-1 event files were reprocessed with CIAO
311: v3.0.1 and CALDB v1.4, using the {\sc acis\_process\_events} tool.
312: No adjustment was made for charge transfer inefficiency (CTI) effects
313: between pixels during the data readout. This allows the analysis of 
314: data to be uniform for different CCD detectors and degrees of CCD pile-up.
315: To minimize pileup effects, we restricted the count rates for on-axis 
316: full-frame (frame time 3.24~s) CCD observations to be   
317: $<$0.08~s$^{-1}$. According to the {\it Chandra} 
318: Proposer's Observatory Guide\footnote{Available at http://cxc.harvard.edu/proposer/POG/}, 
319: this corresponds to 10\% pileup. Count rates in excess of this 
320: value for point sources are likely to impact the extracted spectra. 
321: PHA randomization was applied, but pixel randomization was not.
322: 
323: In cases where the X-ray sources were observed off-axis or in a subarray CCD mode,
324: the pileup effect is reduced, and we can accept a higher net count rate. 
325: The actual pile-up fraction is estimated for the ``reduced'' count rates in 
326: Table~\ref{table1}. These count rates were calculated by taking into account the larger 
327: point-spread function for sources observed off-axis and the CCD observation mode.
328: Column (8) of the table lists the ACIS CCD in which the source is imaged and the subarray value,
329: i.e., the fraction of the CCD used in the observation. Exposing a smaller chip 
330: area results in shorter frame times and reduces the pileup.
331: 
332: Source spectra were typically extracted from regions of radius 2$\arcsec$, 
333: and local background spectra were extracted using annuli 
334: with inner and  outer radii of 6$\arcsec$ and 10$\arcsec$. 
335: For off-axis sources we used elliptical regions, 
336: and for crowded regions, slightly more complicated background 
337: regions, as needed. Visual inspection ensured 
338: that there was no confusion with any nearby X-ray sources.
339: Sources were retained that had $>$1000 counts in the reprocessed data.
340: Spectral fitting was performed using XSPEC v11.2.0bd.
341: 
342: A total of 21 sources had less than 1000 counts after the archival data 
343: were fully reprocessed, so these are rejected. In addition, 
344: 9 observations of 7 sources have $>$10\% pile-up and are rejected 
345: (Table~\ref{table1}). Most of the sources with significant pile-up have
346: other {\it Chandra} observations, so only 3 unique objects are 
347: fully rejected from our sample because of pile-up, 
348: 2 ULXs and one lower luminosity source.
349: 
350: 
351: \subsection{Final Sample}
352: 
353: A total of 69 unique objects in 89 data sets comprise our final sample.
354: The properties of these objects are listed in Table~\ref{table2}, 
355: together with some properties of their host galaxies.
356: Using count rates derived from our reprocessed data, we re-computed
357: the 0.3$-$8.0~keV observed luminosities (L$_X$), using a
358: PL model with $\Gamma=$~1.8 and Galactic absorption. A final 
359: division into two groups is made according to the 
360: maximum observed luminosity, L$_X^{max}$. There are 47 ULXs 
361: (L$_X^{max}$~$\ge$~10$^{39.0}$ erg~s$^{-1}$)
362: and 22 comparison objects of lower-luminosity
363: (L$_X^{max}$~$<$~10$^{39.0}$ erg~s$^{-1}$).
364: 
365: Some sources show luminosity variability. 
366: For two ULXs, U2 (M33~X-8) and U41 (IXO~83), their luminosity can fall below 
367: our threshold value of 10$^{39.0}$ erg~s$^{-1}$ in some cases, but we still
368: retain the classification of ULX.
369: Our method identifies a ULX as such if it is observed with 
370: L$_X$~$\ge$~10$^{39.0}$ erg~s$^{-1}$ at least once.
371: On the other hand, a well-known and previously studied ULX, IXO~85 
372: (C22) is excluded from our ULX sample because the {\it Chandra} 
373: luminosity falls just below our ULX limit.
374: 
375: Examining the galaxy properties in Table~\ref{table2}, we find that most of 
376: our sample objects reside in 
377: spiral or irregular (merger) galaxies and are preferentially located
378: in spiral arms and star-forming regions. Our galaxy sample includes two 
379: mergers (NGC 520 and the Antennae) and four early type galaxies 
380: (NGC 2681, NGC 4125, M87, and Cen A).
381: We see little difference between the ULX locations in their host galaxies
382: in general and the locations of the comparison sources. The two groups
383: also tend to have similar deprojected offsets from the centers of their galaxies.
384: Two ULXs (U2 and U14) are associated with 
385: the nucleus of their host galaxies (M33 and NGC 3310, respectively), 
386: but with no evidence of AGN activity. We do not have enough detailed 
387: information on these sources to know what fraction are known low-mass X-ray 
388: binaries (LMXBs) and high-mass X-ray binaries (HMXBs). 
389: The identification of the optical counterparts would require sensitive 
390: optical imaging. Based on their location in the host galaxy, we can only say that 
391: most of our sources in both samples are consistent with being HMXBs.
392: 
393: 
394: \subsection{Background Contamination}
395: 
396: Here we estimate potential sample contamination from additional 
397: background objects that have not already been clearly identified.  
398: We use the log(N)$-$log(S) function from X-ray deep field surveys to 
399: estimate the fraction of additional background objects based on our 
400: source fluxes and galaxy distances. For our 
401: sample criteria, we construct two flux limits, FL and FC. FL is the flux of 
402: a source with a specific luminosity: 10$^{39.0}$~erg~s$^{-1}$ for ULXs 
403: and 10$^{38.3}$~erg~s$^{-1}$ for our comparison sample. FC is the flux 
404: of a source that provides 1000 counts in its spectrum for the longest 
405: exposure time obtained for each galaxy. Assuming a PL model with 
406: $\Gamma=$~1.8 and using the Galactic value of absorption corresponding
407: to the location of the center of each galaxy on the sky we calculate 
408: FL and FC for all of the 286 galaxies in our original list 
409: (see section 2.1). The final flux limit for each galaxy to compare 
410: with log(N)$-$log(S) is the largest of the two fluxes, FC or FL. 
411: All of our measured fluxes are above 
412: 10$^{-14}$~erg~cm$^2$~s$^{-1}$, which corresponds to an ACIS count rate 
413: of $\sim$10$^{-3}$~s$^{-1}$. 
414: 
415: To make a background estimate we also need to account for the size
416: of the detectors on the sky compared to the projected sizes of the galaxies.
417: The area of each galaxy in deg$^2$ is first calculated within the D$_{25}$ ellipse. 
418: Most observations are done in ACIS imaging mode, 
419: with detector areas of $\sim$0.117~deg$^2$ for both ACIS-I and ACIS-S.
420: Data can be extracted from specific CCD chips, and some observations are only 
421: in subarray mode with a significantly smaller exposed area. The disparity 
422: between the sizes of the galaxies and the detector coverage can 
423: affect our background estimates. For the nearest galaxies, 
424: their size on the sky is larger than or comparable to the size of the ACIS detectors.
425: Naturally, if the projected area of the galaxy is larger than or 
426: comparable to the size of the detector, these galaxies will
427: provide the largest estimated contributions to the background counts.
428: We therefore account for the fractional coverage of the 13 largest galaxies by 
429: over plotting the CCDs and estimating the coverage fraction. These 13 galaxies 
430: (out of the original 286) contribute 65\% to the total estimate 
431: of the contamination. For the remaining galaxies we use the D$_{25}$ ellipse area. 
432: 
433: We used the log(N)$-$log(S) function from two separate surveys to obtain 
434: flux estimates. The popular {\it ROSAT} deep survey 
435: in the Lockman Hole \citep{has98} gives log(N)$-$log(S) 
436: for the flux interval 10$^{-15}-$10$^{-13}$~erg~cm$^2$~s$^{-1}$, 
437: in the range 0.5$-$2~keV. We apply a scale factor of 0.38 for our 
438: 0.3$-$8~keV band, obtained using the absorbed PL model with $\Gamma=$~1.8.
439: The {\it Chandra} Multiwavelength Project (ChaMP) serendipitous survey 
440: \citep{kim04} contains a larger sample, and covers a wide area 
441: ($\sim$14 deg$^2$). It uses the same soft X-ray band as the {\it ROSAT} deep survey,
442: but the slope of the log(N)$-$log(S) function is shallower at the 
443: high end.
444: 
445: For ULXs, the {\it ROSAT} and ChaMP surveys predict no more than 3 or 
446: 5 spurious sources, respectively.
447: For our lower luminosity objects, the prediction is 1 or 2 spurious sources.
448: The survey estimates are compatible given large errors due to 
449: poor sampling at the high flux end. Thus, no more than 
450: approximately one in ten sources in our sample is likely a background object. 
451: In a practical sense this is an upper limit, as our estimate does not 
452: take into account the variable absorption column through each galaxy, which will 
453: attenuate the signal of any background sources shining through the galaxy 
454: (i.e. reduce their observed flux). This is especially important, 
455: as we have used surveys in the 0.5$-$2 keV band where 
456: absorption is strong. We also remind the reader that we have already 
457: identified and rejected two background quasars (Section~2.3).
458: 
459: 
460: 
461: \section{SPECTRAL ANALYSIS}
462: 
463: We grouped the spectra to have a minimum of 15 counts per energy bin
464: for the energy range of 0.3$-$8.0~keV.
465: All fits were performed using the Galactic absorbing column
466: (as listed in Table~\ref{table2}), plus an intrinsic absorbing column 
467: for each galaxy.
468: Galactic values were obtained with the COLDEN routine in CIAO,
469: which provides a foreground N$_H$ value at a given celestial position.
470: We chose to define acceptable (or ``good'') fits as
471: those for which $\chi_{\nu}^{2}\leq~$1.2. Unless specified, all errors quoted
472: are 90\% confidence for one interesting parameter ($\Delta\chi^{2} =$~2.7).
473: For sources with multiple observations, the individual observations were
474: first fitted separately and then all observations were fitted together
475: in XSPEC for the various purposes of our work. Simultaneous fits
476: are used in the histograms and listed in the tables (e.g. Table~\ref{table3})
477: and individual fits are shown in some of the plots to demonstrate
478: any variability in luminosity and spectral shape. For the simultaneous
479: fits, the model parameters were constrained to the same value in XSPEC,
480: and only the normalizations of model components were allowed to vary freely.
481: 
482: In a statistical sense, spectral fitting results can be strongly biased 
483: by the number of counts in each spectrum. To test for such biases between
484: the ULX and the comparison samples, we constructed histograms of
485: net counts in the spectra (see Figure~1). For sources with multiple observations, 
486: we chose the observation that contained the largest number of
487: counts in its spectrum (see Table~\ref{table2}) to represent in the
488: histogram. The distributions of the number of counts for objects in the
489: samples are similar. A Kolmogorov-Smirnov (K-S) test provides a
490: probability of 0.86, indicating that we have no reason to reject the
491: hypothesis that the distributions are identical in count space. Thus,
492: the two samples have equal sensitivity to spectral features for our model 
493: fitting.
494: 
495: 
496: \subsection{Single-Component Spectral Fits}
497: 
498: Recent spectral analysis of {\it Chandra} ULX spectra shows that many 
499: are well fitted by simple models \citep[e.g.][]{hum03,swa04}.
500: We therefore fit all the spectra with either an absorbed
501: PL model or a multi-color disk blackbody (MCD) model,
502: with absorption fixed at the Galactic value in XSPEC.
503: To keep our results within physical bounds, we impose upper limits of
504: $\Gamma\leq$~10 and kT$_{in}\leq$~4~keV, respectively.
505: The results are listed in Table~\ref{table3}.
506: Both the ULX and the comparison samples are generally well
507: fitted by the absorbed PL model (66\% and 50\%, respectively, are good fits, 
508: as indicated in col. [5] of the table). 
509: For the absorbed MCD model, good fits comprise 45\% and 50\% of the samples, respectively.
510: 
511: The histograms in Figure~2 show the distributions of the photon index
512: and inner disk temperature, normalized to allow for easy comparison. 
513: For the full sample, we find no significant difference between
514: ULXs and lower luminosity objects. Luminosity dependences are
515: presented in Figure~3. For objects that have multiple observations,
516: all fit results are shown.
517: 
518: We have applied the K-S test and
519: the T-test to the samples in different ways.
520: The first row of Table~\ref{table4} shows the results of the test applied to the total set of fits,
521: while the second row is restricted to the ``good'' fits as defined
522: in the first paragraph of Section~3. All calculated probabilities
523: are higher than a 5\% significance level, confirming that there
524: are no significant differences when comparing the distributions
525: or their means. We note that the derived probabilities differ in
526: some cases significantly between the K-S test and the T-test,
527: which is an indication that the distributions plotted in Figure~2
528: are possibly derived from intrinsic samples that do not have
529: normal distributions and/or that our sample sizes are small (such
530: tests are usually more reliable when applied to large samples).
531: 
532: 
533: Even with these caveats, we find an interesting trend if we
534: limit our sample further. When only considering the good fits,
535: the disk temperatures are marginally higher for ULXs 
536: (at 1.8~keV, with a significance level of 7\%-8\%). 
537: If we further use Figure~3a to split the ULXs
538: themselves into two groups, with a luminosity break at
539: 5$\times$10$^{39}$~erg~s$^{-1}$, then we find that the
540: highest luminosity ULXs have significantly harder X-ray spectra
541: than both the lower luminosity ULXs and the comparison sample
542: (rows 3-6 of Table~\ref{table4}).
543: 
544: Our primary result from applying single-component models is that
545: all of the highest luminosity ULXs that are well fitted by the PL
546: model possess hard X-ray spectra ($\Gamma\leq$~2 and
547: kT$_{in}\geq$~1.3~keV). The most luminous ULXs have harder spectra,
548: and those that are less luminous have spectral shapes similar to the
549: comparison sample. Not all of the high-luminosity ULXs have
550: hard spectra, however, and so we have further defined a subsample
551: of 9 very luminous and hard ULXs (see Fig.~3a): U4, U5, U10, U11, U14, U18, U19 (with
552: 4 observations), U20, and U43. These all have luminosities in
553: excess of 5$\times$10$^{39}$~erg~s$^{-1}$ and photon indices
554: $<$1.7. This subclass is discussed further in the next sections.
555: 
556: 
557: \subsection{Two-Component Spectral Fits}
558: 
559: We next fit all spectra with the frequently used two-component model 
560: that consists of a MCD model plus a PL. Typical spectral states 
561: observed in black hole binaries and some ULXs \citep[e.g.][]{kub01} include a soft (high) state,
562: with a prominent blackbody component having kT~$\sim$~1~keV plus a steep
563: ($\Gamma\sim$~2.5) PL tail, or a hard (low) state with the thermal
564: component being generally cooler or nonexistent and most of the
565: energy carried in a shallower PL ($\Gamma\sim$~1.8).
566: We also mention the very high state (VHS),
567: characterized by high luminosities, a steep PL ($\Gamma>$~2.5),
568: a relatively cool disk, and sometimes X-ray quasi-periodic oscillations
569: \citep[QPOs; see][for a detailed description]{mcc06}.
570: 
571: We note that these spectral states have been traditionally measured in
572: the 2$-$20~keV energy band and therefore may not be recognized easily
573: in the {\it Chandra} band. For example, in the high state the PL
574: component would be completely absent in our 0.3$-$8~kV band. Also, one of the most
575: important signatures expected from an IMBH is a cool accretion disk component.
576: The inner disk temperature in the MCD model scales with the black hole
577: mass as $\propto M^{-1/4}$. For typical values of kT$_{in}\sim$~1~keV for a black
578: hole binary with $10M_{\odot}$ in the high state, we would expect cool disks
579: with kT$_{in}\sim$~0.1$-$0.3~keV. A number of ULXs with high-quality spectra
580: from {\it Chandra}, {\it XMM-Newton}, and {\it RXTE} were found in the past few years
581: to show soft components well fit by an MCD model in this range \citep[see][]{mill04}.
582: 
583: To compare with published results and restrict model parameters enough to be
584: useful for our purposes, we select a two-component model
585: with fixed parameters. We assume inner disk temperatures of 0.25 or 1~keV
586: to represent either a cool disk or a ``normal'' disk temperature,
587: respectively (models PLMCD0.25 and PLMCD1.0). For ULXs, good fits are derived
588: for 70\% and 72\% of the sample for PLMCD0.25 and PLMCD1.0, respectively.
589: For the comparison sample, good fits are derived for 59\% and 55\% for
590: PLMCD0.25 and PLMCD1.0, respectively. The ULXs do possess a higher
591: percentage of good fits, but the difference is not statistically
592: significant given our sample sizes.
593: Figure~4 shows the distribution of photon indices.
594: The shaded areas correspond to the subsample of 9 ULXs with high luminosities
595: and hard X-ray spectra as defined in the previous section (see Figure~3a).
596: In total, there is no significant difference between the ULXs and the
597: lower-luminosity sources.  A K-S test for the difference 
598: between the distributions gives probabilities of 0.21 and 0.15 
599: for PLMCD0.25 and PLMCD1.0, respectively.
600: However, the 9 high-luminosity, hard ULXs clearly stand out.
601: We note the very steep PL component in some spectra for the PLMCD1.0 model.
602: These results correspond to the ``nonstandard model'' fits of \citet{wil06}.
603: 
604: We tried our two-component model with all parameters free (PLMCD model),
605: but many parameters are not constrained. Moreover, as seen in Figure~5,
606: the MCD component is very weak or practically nonexistent in many cases.
607: The nine high-luminosity, hard ULXs have the weakest disk components,
608: practically negligible.
609: We note a very steep PL component in some spectra here, again indicating
610: a nonstandard model. In these spectra the nonthermal component is soft
611: and strongly absorbed, as shown by the large values of the flux ratios. 
612: Here we only comment further on specific results for spectra that were not well fitted with
613: the simple models from Section 3.1. Table~\ref{table5} presents the PLMCD model 
614: results and in Figure~6 we plot absorbed luminosities
615: versus the photon index and disk temperature. The two samples
616: do not show significant differences. Both samples possess cool disks,
617: and there is no apparent correlation of the disk temperature with luminosity.
618: The presence of this soft disk component also causes the PL slopes
619: to generally become steeper, compared to our single PL fits (Figure~3).
620: 
621: 
622: The use of applying the F-test for an {\it added} spectral component
623: \citep{pro02} is controversial, so we performed simulations to check the validity of the
624: F-tests. For each spectrum we performed 500 simulations under a null model,
625: a PL in this case. We first used the command ``tclout simpars'', available in XSPEC v.12
626: to generate simulated parameters from the original fits.
627: This method uses simulations from a multivariate normal distribution
628: based on the covariance matrix estimated in the original fit.
629: The simulated F-test results are listed in parentheses in column~(7) of Table~\ref{table5}.
630: Any differences between the simulations and the classical F-test
631: are small and generally fall within the errors corresponding to the number
632: of simulations ($\sim$~5\%). 
633: The method described in  \citet{pro02} uses
634: a complete Bayesian Monte Carlo simulation to sample
635: from the posterior distribution \citep[developed by][]{van01}.
636: Our method {\it approximates} the posterior distribution
637: with a multivariate normal distribution centered at the best-fit value.
638: This is nevertheless better than just using the ``fakeit'' command
639: on the original spectrum fitted with the null model \citep[described by]
640: [as a ``parametric bootstrap'', and only valid when the parameters are
641: very well constrained]{pro02}. 
642: 
643: 
644: In conclusion, for the subsample of 9 high-luminosity, hard ULXs (Figure~3a and
645: Section~3.1) we recover the same result here;  they tend to have significantly
646: harder spectra. We also verify that they tend to possess small contributions from
647: a thermal component. If such a component exists, it is practically undetectable 
648: with the {\it Chandra} data.
649: We also find that cool disks (MCD with kT$_{in}\sim$~0.1$-$0.3~keV)
650: are present with roughly equal probability for any given ULX
651: and that cool-disk components extend below the standard ULX luminosity cutoff (10$^{39}$~erg~s$^{-1}$),
652: down to our sample limit of 10$^{38.3}$~erg~s$^{-1}$.
653: 
654: 
655: \section{SHORT-TERM VARIABILITY}
656: 
657: Long-term flux variability from one observation to another,
658: which is typically years, is very common in ULXs. Short-term
659: flux variability, which we define here as that which can be detected
660: within a single observation (hours), is less frequent and is not easily
661: found with {\it Chandra}, probably due to limited sensitivity (i.e., not providing enough counts).
662: Using the K-S test, \citet{swa04} find that $\approx$15\% of
663: our ULXs are variable at the 95\% confidence level.
664: 
665: 
666: We extracted light curves for all sources, using three time bins:
667: 3.24 (nominal frame time), 500, and 1000~s. To test for variability,
668: we used the K-S statistic for the nominal frame time binning, 
669: and the Chi-Squared test for the other two.
670: Using the Monte Carlo method of \citet{par06} described in the previous section,
671: we constructed light curves for the hardness ratios for each variable source 
672: and looked for variations in hardness ratios and possible time lags between
673: the three energy bands. We also constructed power spectra using the Leahy
674: normalization.
675: 
676: We detect variability at 95\% confidence in 6 ULXs for the longer time
677: frames using the Chi-Squared test, and no variability for the lower luminosity sample.
678: Of these, three sources were previously known to be variable. These are U2, U34, and U40.
679: Three other sources show variability. These are U14, U27 and C22, and the
680: variability scale is similar to the exposure times of the observations ($\sim$40 ks).
681: The K-S test identifies the same variable sources with the exception of U33 in NGC 5055,
682: but it finds significant variability in two additional sources:
683: U27 in NGC 4565 and a comparison source, C22 in NGC 6946.
684: There are two periodicities of 707~s detected in U33 and U6 (NGC 1313 X-1)
685: produced by the ACIS dither, which causes false periodic signals
686: at 707 and 1000~s.
687: 
688: 
689: We conclude that  6 ULXs are intrinsically variable, which is consistent
690: with the result obtained by \citet{swa04} given the small size of the sample (47 sources).
691: Only two sources (U2 and U40) show some energy variation, but no lag.
692: Given the readout time of the {\it Chandra} CCDs,
693: variations on timescales shorter than $\approx$10~s cannot be detected,
694: and features that could identify spectral states (QPOs or breaks in the power
695: density spectra), are not readily detectable.
696: 
697: 
698: 
699: \section{DISCUSSION}
700: 
701: From our X-ray spectral comparison between ultraluminous X-ray sources (ULXs) 
702: and other X-ray point sources in nearby galaxies, we find an interesting subclass of nine ULXs 
703: that have unique properties compared to the other sources that are classified as ULXs. 
704: This subclass of ULXs also differs from the lower-luminosity sample of X-ray point sources. 
705: While most of the ULXs we analyzed can plausibly be explained as scaled-up versions 
706: of Galactic black hole binaries, this particular subclass cannot. 
707: We discuss this subclass of ULXs followed by our general results, 
708: especially how our results relate to current evidence that supports the idea 
709: that ULXs host intermediate-mass black holes (IMBHs).
710: 
711: \subsection{Luminous, Hard (flat-spectrum) X-ray ULXs}
712: 
713: Our analysis has identified nine ULXs with very high luminosities and hard (flat) X-ray spectra. 
714: This sample (U4, U5, U10, U11, U14, U18, U19, U20, and U43) is shown 
715: in Figure~3a (upper left corner). These ULXs are all well fitted 
716: by a power-law (PL) model with a photon index of $<$1.7. More complex spectral fitting 
717: using two component models reveals that when trying to add an accretion disk component 
718: (the MCD model described above), the contribution of this component 
719: to the total X-ray flux is very small, indicating that the relative contribution 
720: of emission from the accretion disk to the total spectrum is small. 
721: This result is shown in Figure~5, where the ratio of the absorbed MCD component flux 
722: to the total flux is plotted against the PL index for the PLMCD model. 
723: By using absorbed fluxes we do not specifically show the absolute physical strength 
724: of the accretion disk component; however, the ratios measured in this way 
725: better indicate the significance of detecting a soft excess 
726: and are also less dependent on the modeling.
727: 
728: The nine members of the subclass of flat-spectrum ULXs are also very luminous, 
729: so they would seem to be the best candidates for hosting an IMBH 
730: based on a simple Eddington limit argument, which predicates 
731: that higher mass black holes are required to explain the most luminous accreting sources. 
732: The spectra of these ULXs resemble Galactic black holes in a hard state, 
733: but such spectral shapes are usually associated with a low-luminosity state 
734: in the case of Galactic black holes. If these ULXs are indeed accreting IMBHs 
735: in a low state (i.e. low/hard state), our result begs the question 
736: as to why we do not also see ULXs in a high state (high/soft state) 
737: with even higher luminosities. Moreover, if these are IMBHs in a low state, 
738: such a scenario implies very high mass black holes ($>$10$^4$~M$_{\odot}$). 
739: The formation of such black holes is not easy to explain.
740: 
741: It seems more plausible that this subclass of hard (flat) and luminous ULXs 
742: are accreting sources in the PL dominated very high state \citep[VHS;][]{mcc06}, 
743: with an unusually weak soft X-ray component. A model that describes the properties 
744: of a hard PL with very little flux from the disk, at least in the {\it Chandra} band, 
745: is the ``coupled disk-corona'' model proposed by \citet{don06}. In this model, 
746: the underlying accretion disk emission is distorted by a process 
747: that drains energy from the disk into the corona. In an extreme case, 
748: the inner disk emission could be almost completely Comptonized, 
749: and thus only the visible outer disk would contribute to the accretion disk component.  
750: As this obviously only appears at low temperatures it could be easily absorbed 
751: in some galaxies and also hard to detect. However, if our nine luminous ULXs 
752: are interpreted as stellar-mass black hole systems in a high state, 
753: we would need to explain why their X-ray spectra are much harder 
754: compared to those observed in our Galaxy \citep{mcc06}, 
755: which have typical photon indices of ~2.5 in the VHS. 
756: We would also need to explain how such low-mass black hole systems 
757: could reach such high luminosities.
758: A model that could explain both the flat non-thermal component 
759: and the weak soft disk component in luminous ULXs has been proposed by \citet{soc06}.
760: Their ULX model shows that at super-Eddington accretion rates, 
761: in the inner region of the disk, magnetic fields in the corona can prevent strong winds, 
762: thus the radiative efficiency is not reduced by photon trapping. 
763: The resulting spectrum is dominated by the coronal emission from the inner region 
764: and the soft thermal component is generated only in the outer disk.
765: 
766: 
767: There is some direct evidence that high-luminosity states in ULXs 
768: correlate with the hardness of the PL tail in their spectra. 
769: \citet{rob06} have shown this for a long {\it Chandra} observing campaign of NGC~5204 X-1 (our U36). 
770: These data were not available when we searched the archive. 
771: \citet{rob06} found that the spectrum becomes harder as the flux increases, 
772: over time-scales of days to weeks. The model used was a Comptonized disk model, 
773: and the results showed a cool disk ($\sim$0.1~keV) and an optically thick corona. 
774: This model demonstrates that flux variations correlate with the corona temperature. 
775: \citet{rob06} favor a stellar mass black hole interpretation for this ULX 
776: and suggest an unusual VHS, probably produced by extreme mass transfer 
777: from a massive star. We should note, however, that the PL slope is much steeper 
778: for NGC~5204 X-1 compared to the ULXs we discuss; therefore it is much easier to interpret 
779: as a VHS in comparison to what is observed for Galactic black holes.
780: 
781: It is possible that our nine ULX spectra appear to be harder than they actually are 
782: due to the limited energy band covered by {\it Chandra}. If these objects intrinsically possess 
783: a break or curvature in their spectra and the break occurs at an energy 
784: above the {\it Chandra} bandpass or where the sensitivity of {\it Chandra} 
785: falls off significantly, this might bias our modeling to measuring these sources as ``hard''. 
786: This appears to be the case for one ULX in our sample, NGC 1313 X-1 (U5). 
787: Using {\it XMM-Newton} spectra, \citet{sto06} found evidence for a break (or curvature) at 4.9~keV, 
788: with a photon index for the high-energy PL of 2.16 (much closer to Galactic black holes values). 
789: The authors show, however, that such breaks are easier to explain 
790: if ULXs contain stellar mass black holes rather than IMBHs. The curvature would be likely 
791: to originate in optically thick coronae. This theory would need to be tested 
792: for the remaining ULXs in our subclass by obtaining better quality spectra.
793: 
794: In conclusion, the subclass of nine ULXs with very high luminosities and hard (flat) X-ray spectra
795: suggests a PL-dominated VHS, in line with recent models of stellar mass black hole systems
796: in very high accretion states. The fact that the highest luminosity ULXs are explained more easily
797: with such models argues strongly against IMBHs as the only explanation of ULXs.
798: 
799: 
800: 
801: \subsection{Cool disks and the IMBH interpretation}
802: 
803: Using our sample of ULXs and lower luminosity X-ray sources, we have found 
804: that the spectral signature of a cool accretion disc is not specific to ULXs. 
805: The results of the widely used PLMCD model (see Fig.~6 and Table~\ref{table5}) 
806: show that many sources in both samples have MCD components with low inner-disk temperatures. 
807: Cool disks have been used until recently as support for the IMBH interpretation. 
808: Our results show that this evidence is nowhere near as compelling when a large sample of ULXs 
809: is looked at properly.
810: 
811: In the standard accretion disk model, cool disks are not expected for stellar-mass black holes 
812: accreting near their Eddington limits. The disk temperature scales with the black hole mass as
813: T$_{in}\propto$~M$^{-1/4}$, and is $\sim$~1~keV for stellar mass black holes. 
814: However, cool disks can be seen in a low (hard) state, because the temperature dependence 
815: on the accretion rate for standard disks is T$_{in}\propto~\dot{M}^{-1/4}$ \citep{mcc06,mill06}. 
816: Cool disks have indeed been found in some non-ULX sources \citep[e.g.][]{sto06}. 
817: The authors note the similarity of these spectral fits with those typically used for ULXs. 
818: They also suggest that the soft excess in some cases could be otherwise explained 
819: by contamination from the host galaxy.
820: 
821: Most sources in Figure~6b possess low disk temperatures, within both the ULX and comparison samples. 
822: Indeed, it is surprising that we do not see many states that are typical (high) states 
823: for stellar mass black hole binaries, with a prominent $\sim$1~keV blackbody component 
824: (although a hard PL tail would be difficult to discern in the limited {\it Chandra} bandpass). 
825: Only U2, U37, C3 and C9 show such high temperatures. We find a similar result 
826: from our cool-disk model (PLMCD0.25). A large fraction of our ULX spectra (70\%) 
827: are well fitted by this model, but a significant number of lower luminosity objects (59\%) are as well.
828: 
829: There are theoretical models that do not require the presence of an IMBH 
830: to explain a cool disk at high accretion rates and high luminosities (i.e., for ULXs). 
831: We already mentioned the model proposed by \citet{don06}, that explains cool disks 
832: by a process of draining energy from the disk to launch an optically-thick corona 
833: that obscures the hot inner regions of the disk. \citet{fre06} developed microblazar models 
834: with magnetized jets that cause a transference of disk energy into the jet, 
835: thus making the disk fainter and cooling it at the same time. 
836: Other phenomenological models include the ``dual thermal'' model of \citet{wil06}, 
837: in which the soft excess comes from an optically thick outflow 
838: produced at high accretion rates \citep[see][]{king03}, which is seen 
839: in addition to a disk component with a temperature similar to those seen 
840: in stellar-mass black hole binaries. 
841: This latter model was proposed to explain the alternate model of \citet{wil06}.
842: 
843: \citet{don06} found evidence of a Galactic black hole that supports the interpretation of ULXs 
844: as stellar mass black holes in a VHS. The microquasar XTE J1550-564 has a ``strong VHS'' 
845: \citep[see also][]{kub04}, where the disc temperature decreases with luminosity, 
846: reaching values of 0.3$-$0.4~keV. 
847: A similar behavior has been found for NGC~1313 X-2 \citep{fen07}.
848: This suggests a new type of VHS, 
849: a so-called ultraluminous branch, which is very similar to the ULX spectra
850: \citep{rob07,sor07}. In this interpretation, ULXs represent the high end 
851: of such an accretion state, with black hole masses up to 100~M$_{\odot}$ and accretion rates 
852: up to 20 times the Eddington limit. Forming black holes with such masses 
853: is much easier to explain than forming IMBHs. For example \citet{bel04} 
854: showed that black holes with masses of 80~M$_{\odot}$ or more can form through binary mergers. 
855: \citet{sorw06} suggested that black holes of up to 200~M$_{\odot}$ could form 
856: by large-scale dynamical collapse of protoclusters in active regions in galaxies. 
857: These formation mechanisms are supported by the association between ULXs, 
858: star-forming regions and colliding galaxies.
859: 
860: 
861: 
862: \subsection{Conclusion}
863: 
864: We have found that the highest luminosity ULXs tend to have the hardest X-ray spectra 
865: in the {\it Chandra} bandpass and are well fitted by a simple power-law model, 
866: without evidence for thermal accretion disc components.  Such spectra 
867: are not consistent with current IMBH models, but are more in line with current models 
868: of extreme very high states, or perhaps a new ``ultraluminous state'' \citep{rob07}, 
869: in stellar mass black holes.
870: 
871: Our work shows that cool accretion disks are not exclusive to the ULX 
872: class, suggesting that low-temperature IMBHs are 
873: not the only explanation for this phenomenon.  In general, our results 
874: show that ULXs are likely to be composed of several distinct types of 
875: objects, and that these types may extend into lower X-ray luminosity 
876: classes, such as classical Galactic black hole candidates and other 
877: objects in our comparison sample.  Our conclusions provide another ``nail 
878: in the coffin'' for assumptions that ULXs are simply a class of accreting 
879: IMBHs.
880: 
881: No other specific properties have been found for the ULX group,
882: except for spectral hardening at the highest luminosities. 
883: All these results suggest that ULXs are the highest luminosity end of
884: stellar mass black hole binaries, with the largest black holes permitted by current 
885: formation mechanisms and/or accreting at super-Eddington rates.
886: 
887: 
888: 
889: \acknowledgments
890: 
891: This research has made
892: use of the SAOImage DS9, developed by Smithsonian Astrophysical
893: Observatory and of the NASA/IPAC Extragalactic Database (NED)
894: which is operated by the Jet Propulsion Laboratory, California
895: Institute of Technology, under contract with the National Aeronautics
896: and Space Administration.
897: 
898: 
899: \begin{thebibliography}{}
900: \bibitem[Argyle \& Eldridge(1990)]{arg90} Argyle, R.~W., \& Eldridge, P. 1990, \mnras, 243, 504
901: \bibitem[Barnard et al.(2003)]{bar03} Barnard, R., Kolb, U., \& Osborne, J.~P. 2003, \aap, 411, 553
902: \bibitem[Bauer et al.(2001)]{bau01} Bauer, F.~E., Brandt, W.~N., Sambruna, R.~M., Chartas, G., Garmire, G.~P., Kaspi, S., \& Netzer, H. 2001, \aj, 122, 182
903: \bibitem[Begelman(2002)]{beg02} Begelman, M.~C. 2002, \apj, 568, L97
904: \bibitem[Begelman et al.(2006)]{beg06} Begelman, M.~C., King, A.~R., \& Pringle, J.~E. 2006, \mnras, 370, 399
905: \bibitem[Belczynski et al.(2004)]{bel04} Belczynski, K., Sadowski, A., \& Rasio, F.~A. 2004, \apj, 611, 1068 
906: \bibitem[Bietenholz, Bartel \& Rupen(2002)]{bie02} Bietenholz, M.~F., Bartel, N., \& Rupen, M.~P. 2002, \apj, 581, 1132
907: \bibitem[Bryant \& Hunstead(1999)]{bry99} Bryant, J.~J., \& Hunstead, R.~W. 1999, \mnras, 308, 431
908: \bibitem[Clark et al.(2005)]{cla05} Clark, D.~M., et al. 2005, \apj, 631, L109
909: \bibitem[Colbert et al.(2004)]{col04} Colbert E., Heckman T., Ptak A., Strickland D., \& Weaver K.~A., 2004, \apj, 602, 231
910: \bibitem[Colbert \& Mushotzky(1999)]{col99} Colbert, E.~J.~M., \& Mushotzky, R.~F. 1999, \apj, 519, 89
911: \bibitem[Colbert et al.(1995)]{col95} Colbert, E.~J.~M., Petre, R., Schlegel, E.~M., \& Ryder, S.~D. 1995, \apj, 446, 177
912: \bibitem[Colbert \& Ptak(2002)]{col02} Colbert, E.~J.~M., \& Ptak, A.~F. 2002, \apjs, 143, 25
913: \bibitem[Cropper et al.(2004)]{cro04} Cropper, M., Soria, R., Mushotzky, R.~F., Wu, K., Markwardt, C.~B., \& Pakull, M. 2004, \mnras, 349, 39
914: \bibitem[de Vaucouleurs et al.(1991)]{dev91} de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H.~G., Buta, R.~J., Paturel, G., \& Fouque, P. 1991, Third Reference Catalog of Bright   Galaxies (RC3, New York: Springer-Verlag)
915: \bibitem[Done \& Kubota(2006)]{don06} Done, C., \& Kubota, A. 2006, \mnras, 371, 1216
916: \bibitem[Dubus \& Rutledge(2002)]{dub02} Dubus, G., \& Rutledge, R. 2002, \mnras, 336, 901
917: \bibitem[Eracleous et al.(2002)]{era02} Eracleous, M., Shields, J.~C., Chartas, G., \& Moran, E. C. 2002, \apj, 565, 108
918: \bibitem[Fabbiano et al.(2001)]{fab01} Fabbiano, G., Zezas, A., \& Murray, S.~S.  2001, \apj, 554, 1035
919: \bibitem[Fender \& Belloni(2004)]{fen04} Fender, R., \& Belloni, T. 2004, \araa, 42, 317
920: \bibitem[Feng \& Kaaret(2005)]{fen05} Feng, H., \& Kaaret, P. 2005, \apj, 633, 1052
921: \bibitem[Feng \& Kaaret(2007)]{fen07} Feng, H., \& Kaaret, P.\ 2007, \apjl, 660, L113
922: \bibitem[Foschini et al.(2004)]{fos04} Foschini, L., Rodriguez, J., Fuchs, Y., Ho, L.~C., Dadina, M., Di Cocco, G., Courvoisier, T.~J.-L., \& Malaguti, G.  2004, \aap, 416, 529
923: \bibitem[Freeland et al.(2006)]{fre06} Freeland, M., Kuncic, Z., Soria, R., \& Bicknell, G.~V. 2006, \mnras, 372, 630
924: \bibitem[Gao et al.(2003)]{gao03} Gao, Y., Wang, Q.~D., Appleton, P.~N., \& Lucas, R.~A.\ 2003, \apjl, 596, L171 
925: \bibitem[Ghosh et al.(2006)]{gho06} Ghosh, K.~K., Finger, M.~H., Swartz, D.~A., Tennant, A.~F., \& Wu, K. 2006, \apj, 640, 459
926: \bibitem[Goad et al.(2006)]{goad06} Goad, M.~R., Roberts, T.~P., Reeves, J.~N., \& Uttley, P.\ 2006, \mnras, 365, 191 
927: \bibitem[Gon\c{c}alves \& Soria(2006)]{gon06} Gon\c{c}alves, A.~C. \& Soria R. 2006, \mnras, 371, 673		     
928: \bibitem[Guainazzi et al.(2000)]{gua00} Guainazzi, M., Matt, G., Brandt, W.~N., Antonelli, L.~A., Barr, P., \& Bassani, L. 2000, \aap, 356, 463
929: \bibitem[Hasinger et al.(1998)]{has98} Hasinger, G., Burg, R., Giacconi, R., Schmidt, M., Trumper, J., \& Zamorani, G.\ 1998, \aap, 329, 482 
930: \bibitem[Ho et al.(1997)]{ho97} Ho, L.~C., Filippenko, A.~V., \& Sargent W.~L.~W. 1997, \apjs, 112, 315
931: \bibitem[Holt et al.(2003)]{hol03} Holt, S.~S., Schlegel, E.~M., Hwang, U., \& Petre, R. 2003, \apj, 588, 792
932: \bibitem[Humphrey et al.(2003)]{hum03} Humphrey, P.~J., Fabbiano, G., Elvis, M., Church, M. J., \& Balucinska-Church, M. 2003, \mnras, 344, 134
933: \bibitem[Immler et al.(1999)]{imm99} Immler, S., Vogler, A., Ehle, M., \& Pietsch, W. 1999, \aap, 352, 415
934: \bibitem[Jenkins et al.(2004a)]{jen04a} Jenkins, L.~P., Roberts, T.~P., Ward, M.~J., \& Zezas, A.  2004a, \mnras, 352, 1335
935: \bibitem[Jenkins et al.(2004b)]{jen04b} Jenkins, L.~P., Roberts, T.~P., Warwick, R.~S., Kilgard, R.~E., \& Ward, M. J. 2004b, \mnras, 349, 404
936: \bibitem[Kaaret(2002)]{kaa02} Kaaret, P. 2002, \apj, 578, 114
937: \bibitem[Kim et al.(2004)]{kim04} Kim, D.-W. et al. 2004, \apj, 600, 59
938: \bibitem[King \& Pounds(2003)]{king03} King, A.~R., \& Pounds, K.~A. 2003, \mnras, 345, 657	
939: \bibitem[King et al.(2001)]{king01} King, A.~R., Davies, M.~B., Ward, M.~J., Fabbiano, G., \& Elvis, M. 2001, \apj, 552, L109
940: \bibitem[Kong(2003)]{kong03} Kong, A.~K.~H. 2003, \mnras, 346, 265
941: \bibitem[Kong \& Di~Stefano(2005)]{kong05} Kong, A.~K.~H., \& Di~Stefano, R. 2005, \apj, 632, L107
942: \bibitem[Kong et al.(2002)]{kong02} Kong, A.~K.~H., Garcia, M.~R., Primini, F.~A., Murray, S.~S., Di Stefano, R., \& McClintock, J.~E. 2002, \apj, 577, 738
943: \bibitem[Kraft et al.(2001)]{kra01} Kraft, R.~P., Kregenow, J.~M., Forman, W.~R., Jones, C., \& Murray, S.~S. 2001, \apj, 560, 675
944: \bibitem[Kubota et al.(2001)]{kub01} Kubota, A., Mizuno, T., Makishima, K., Fukazawa, Y., Kotoku, J., Ohnishi, T., \& Tashiro, M. 2001, \apj, 547, L119		    
945: \bibitem[Kubota \& Done(2004)]{kub04} Kubota, A., \& Done, C.\ 2004, \mnras, 353, 980 
946: \bibitem[Kuntz et al.(2005)]{kun05} Kuntz, K.~D., Gruendl, R.~A.; Chu, Y.-H., Chen, C.-H.~R., Still, M., Mukai, K., \& Mushotzky, R.~F. 2005, \apj, 620, L31
947: \bibitem[La Parola et al.(2003)]{lapa03} La Parola, V., Damiani, F., Fabbiano, G., \& Peres, G. 2003, \apj, 583, L758
948: \bibitem[Lira, Lawrence \& Johnson(2000)]{lira00} Lira, P., Lawrence, A., \& Johnson, R.~A. 2000, \mnras, 319, 17
949: \bibitem[Lira et al.(2002)]{lira02} Lira, P., Ward, M., Zezas, A., Alonso-Herrero, A., \& Ueno, S.\ 2002, \mnras, 330, 259 
950: \bibitem[Liu \& Bregman(2005)]{liub05} Liu, J.-F., \& Bregman, J.~N. 2005, \apjs, 157, 59
951: \bibitem[Liu \& Mirabel(2005)]{lium05} Liu, Q.~Z. \& Mirabel, I.~F., 2005, \aap, 429, 1125
952: \bibitem[Ma et al.(1998)]{ma98} Ma, C., et al. 1998, \aj, 116, 516
953: \bibitem[Matonick, \& Fesen(1997)]{mat97} Matonick, D.~M., \& Fesen, R.~A., 1997, \apjs, 112, 49
954: \bibitem[Matsumoto et al.(2001)]{mat01} Matsumoto, H., Tsuru, T.~G., Koyama, K., Awaki, H., Canizares, C.~R., Kawai, N., Matsushita, S., \& Kawabe, R. 2001, \apj, 547, 25
955: \bibitem[McClintock \& Remillard(2006)]{mcc06} McClintock J.~E., \& Remillard R.~A., 2006, ``Black Hole Binaries'', in Compact Stellar X-ray Sources, eds. W.H.G. Lewin and M. van der Klis, Cambridge University Press, Cambridge
956: \bibitem[Miller et al.(2004)]{mill04} Miller, J.~M., Fabian, A.~C., \& Miller, M.~C. 2004, \apj, 614, L117
957: \bibitem[Miller et al.(2006)]{mill06} Miller, J.~M., Homan, J., Steeghs, D., Rupen, M., Hunstead, R.~W., Wijnands, R., Charles, P.~A., \& Fabian, A.~C.\ 2006, \apj, 653, 525 
958: \bibitem[Mucciarelli et al.(2006)]{muc06} Mucciarelli, P., Casella, P., Belloni, T.,  Zampieri, L., \& Ranalli P. 2006, \mnras, 365, 1123
959: \bibitem[Mukai et al.(2003)]{muk03} Mukai, K., Pence, W.~D., Snowden, S.~L., \& Kuntz, K.~D. 2003, \apj, 582, 184
960: \bibitem[Mukai et al.(2005)]{muk05} Mukai K., Still M., Corbet R.~H.~D., Kuntz K.~D., \& Barnard R. 2005, \apj, 634, 1085
961: \bibitem[Park et al.(2006)]{par06} Park, T., Kashyap, V.~L., Siemiginowska, A., van Dyk, D.~A., Zezas, A., Heinke, C., \& Wargelin, B.~J. 2006, ApJ, 652, 610
962: \bibitem[Pence et al.(2001)]{pen01} Pence, W.~D., Snowden, S.~L., Mukai, K., \& Kuntz, K.~D. 2001, \apj, 561, 189
963: \bibitem[Pietsch et al.(2001)]{pie01} Pietsch, W. et al. 2001, \aap, 365, L174
964: \bibitem[Poutanen et al.(2007)]{pou07} Poutanen, J., Lipunova, G., Fabrika, S., Butkevich, A.~G., \& Abolmasov, P.\ 2007, \mnras, 377, 1187 
965: \bibitem[Primini, Forman \& Jones(1993)]{pri93} Primini, F.~A., Forman, W., \& Jones, C. 1993, \apj, 410, 615
966: \bibitem[Protassov et al.(2002)]{pro02} Protassov, R., van Dyk, D.~A., Connors, A., Kashyap, V.~L., \& Siemiginowska, A. 2002, \apj, 571, 545
967: \bibitem[Read(2005)]{read05} Read, A.~M. 2005, \mnras, 359, 455
968: \bibitem[Read et al.(1997)]{read97} Read, A.~M., Ponman, T.~J., \& Strickland, D.~K. 1997, \mnras, 286, 626
969: \bibitem[Revnivtsev et al.(2002)]{rev02} Revnivtsev, M., Sunyaev, R., Gilfanov, M., \& Churazov, E.\ 2002, \aap, 385, 904
970: \bibitem[Roberts(1997)]{rob97} Roberts, T.~P., 1997, PhD thesis, Univ. Leicester
971: \bibitem[Roberts \& Colbert(2003)]{roco03} Roberts, T.~P., \& Colbert, E.~J.~M. 2003, \mnras, 341, L49
972: \bibitem[Roberts et al.(2003)]{rob03} Roberts, T.~P., Goad, M.~R., Ward, M.~J., \& Warwick, R.~S. 2003, \mnras, 342, 709
973: \bibitem[Roberts et al.(2001)]{rob01} Roberts, T.~P., Goad, M.~R., Ward, M.~J., Warwick, R.~S., O'Brien, P.~T., Lira, P., \& Hands, A.~D.~P.  2001, \mnras, 325, L7
974: \bibitem[Roberts \& Warwick(2000)]{rob2000} Roberts, T. P., \& Warwick, R. S. 2000, \mnras, 315, 98
975: \bibitem[Roberts et al.(2004)]{rob04} Roberts, T.~P., Warwick R.~S., Ward M.~J., \& Goad M.~R. 2004, \mnras, 349, 1193
976: \bibitem[Roberts et al.(2005)]{rob05} Roberts, T.~P., Warwick, R.~S., Ward, M.~J., Goad, M.~R., \& Jenkins, L.~P.  2005, \mnras, 357, 1363
977: \bibitem[Roberts et al.(2006)]{rob06} Roberts, T.~P., Kilgard, R.~E., Warwick, R.~S., Goad, M.~R., \& Ward, M.~J.\ 2006, \mnras, 371, 1877 
978: \bibitem[Roberts et al.(2002)]{rob02} Roberts, T.~P., Warwick, R.~S., Ward, M.~J., \& Murray, S.~S.  2002, \mnras, 337, 677
979: \bibitem[Roberts(2007)]{rob07} Roberts, T.~P.\ 2007, \apss, 311, 203
980: \bibitem[Schlegel et al.(1997)]{sch97} Schlegel, E.~M., Barrett, P., \& Singh, K.~P. 1997, \aj, 113, 1296
981: \bibitem[Schlegel \& Pannuti(2003)]{sch03} Schlegel, E. M., \& Pannuti, T. G. 2003, \aj, 125, 3025
982: \bibitem[Smith \& Wilson(2001)]{smi01} Smith, D.~A., \& Wilson, A.~S. 2001, \apj, 557, 180
983: \bibitem[Smith \& Wilson(2003)]{smi03} Smith, D.~A., \& Wilson, A.~S. 2003, \apj, 591, 138
984: \bibitem[Socrates \& Davis(2006)]{soc06} Socrates, A., \& Davis, S.~W.\ 2006, \apj, 651, 1049
985: \bibitem[Soria \& Wu(2003)]{sor03} Soria, R., \& Wu, K. 2003, \aap, 410, 53
986: \bibitem[Soria \& Wong(2006)]{sorw06} Soria, R., \&  Wong, D.~S. 2006, \mnras, 372, 1531
987: \bibitem[Soria et al.(2006)]{sor06} Soria, R., Kuncic, Z., Broderick, J.~W., \& Ryder, S.~D.\ 2006, \mnras, 370, 1666 
988: \bibitem[Soria et al.(2007)]{sor07} Soria, R., Baldi, A., Risaliti, G., Fabbiano, G., King, A., La Parola, V., \& Zezas, A.\ 2007, \mnras, 379, 1313
989: \bibitem[Stobbart et al.(2004)]{sto04} Stobbart, A.-M., Roberts, T.~P., \& Warwick, R.~S. 2004, \mnras, 351, 1063
990: \bibitem[Stobbart et al.(2006a)]{sto06} Stobbart, A.-M., Roberts, T.~P., \& Warwick, R.~S., 2006a, \mnras, 370, 25
991: \bibitem[Stobbart et al.(2006b)]{wil06} Stobbart, A-M., Roberts, T.~P., \& Wilms, J. 2006b, \mnras, 368, 397
992: \bibitem[Strickland et al.(2001)]{str01} Strickland, D.~K., Colbert, E.~J.~M., Heckman, T.~M., Weaver, K.~A., Dahlem, M., \& Stevens, I.~R. 2001, \apj, 560, 707
993: \bibitem[Strohmayer \& Mushotzky(2003)]{stro03} Strohmayer, T.~E., \& Mushotzky, R.~F. 2003, \apj, 586, 61
994: \bibitem[Sugiho et al.(2001)]{sug01} Sugiho, M., Kotoku, J., Makishima, K., Kubota, A., Mizuno, T., Fukazawa, Y., \& Tashiro, M. 2001, \apj, 561, 73
995: \bibitem[Summers et al.(2003)]{sum03} Summers, L.~K., Stevens, I.~R., Strickland, D.~K., \& Heckman, T.~M. 2003, \mnras, 342, 690
996: \bibitem[Swartz et al.(2003)]{swa03} Swartz, D.~A., Ghosh, K.~K., McCollough, M.~L., Pannuti, T.~G., Tennant, A.~F., \& Wu, K. 2003, \apjs, 144, 213
997: \bibitem[Swartz et al.(2004)]{swa04} Swartz, D., Ghosh, K. Tennant, A., \& Wu, K. 2004, \apjs, 154, 519
998: \bibitem[Tanaka et al.(2005)]{tan05} Tanaka, T., Sugiho, M., Kubota, A., Makishima, K., \& Takahashi, T. 2005, PASJ, 57, 507
999: \bibitem[Tully(1988)]{tul88} Tully, R. B. 1988, Nearby Galaxy Catalog (Cambridge)
1000: \bibitem[van Dyk et al.(2001)]{van01} van Dyk, D.~A., Connors, A., Kashyap, V.~L., \& Siemiginowska, A. 2001, \apj, 548, 224
1001: \bibitem[V\'eron-Cetty \& V\'eron(2003)]{ver03} V\'eron-Cetty, M.-P., \& V\'eron, P. 2003, \aap, 412, 399
1002: \bibitem[Vogler \& Pietsch(1999)]{vog99} Vogler, A., \& Pietsch, W. 1999, \aap, 342, 101
1003: \bibitem[Vogler, Pietsch \& Bertoldi(1997)]{vog97} Vogler, A., Pietsch, W., \& Bertoldi, F. 1997, \aap, 318, 768
1004: \bibitem[Wang et al.(2003)]{wang03} Wang, Q.~D. , Chaves, T., \& Irwin, J.~A. 2003, \apj, 598, 969
1005: \bibitem[Wang et al.(1999)]{wang99} Wang, Q.~D., Immler, S., \& Pietsch, W. 1999, \apj, 523, 121
1006: \bibitem[Weisskopf et al.(2003)]{wei03} Weisskopf, M.~C., et al. 2003, ExA, 16, 1
1007: \bibitem[Weisskopf et al.(2004)]{wei04} Weisskopf, M.~C., Wu, K., Tennant, A.~F., Swartz, D.~A., \& Ghosh, K.~K. 2004, \apj, 605, 360
1008: \bibitem[Williams et al.(2004)]{wil04} Williams, B.~F., Garcia, M.~R., Kong, A.~K.~H., Primini, F.~A., King, A.~R., Di~Stefano, R. \& Murray, S.~S.  2004, \apj, 609, 735
1009: \bibitem[Winter et al.(2006)]{win06} Winter, L.~M., Mushotzky, R.~F., \& Reynolds, C.~S. 2006, \apj, 649, 730
1010: \bibitem[Wu et al.(2002)]{wu02} Wu, H., Xue, S.~J., Xia, X.~Y., Deng, Z.~G., \& Mao, S. 2002, \apj, 576, 738
1011: \bibitem[Zacharias et al.(2004)]{zac04} Zacharias, N., Urban, S.~E., Zacharias, M.~I., Wycoff, G.~L., Hall, D.~M., Monet, D.~G., \& Rafferty, T.~J.  2004, \aj, 127, 3043
1012: \bibitem[Zezas et al.(2002)]{zez02} Zezas, A., Fabbiano, G., Rots, A.~H., \& Murray, S.~S. 2002, \apjs, 142, 239
1013: %Lira and Gao are quoted directly
1014: 
1015: \end{thebibliography}
1016: 
1017: 
1018: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Table 1 %%%%%%%%%%%%%%%%%   
1019: 
1020: \clearpage
1021: \pagestyle{empty} 
1022: \begin{deluxetable}{r|cccccrrccccll}
1023: \tablecolumns{13}                                    
1024: \rotate                                     
1025: \tablewidth{0pt} 
1026: \tabletypesize{\tiny} 
1027: \tablenum{1}              
1028: \tablecaption{Piled-up observations\label{table1}}          
1029: \tablehead{
1030: \colhead{Obs.} & \colhead{Name} & \colhead{Position} & \colhead{Galaxy} &
1031: \colhead{Log L$_X$} & \colhead{OBSID} & \colhead{Date} &  
1032: \colhead{CCD} & \colhead{$\theta_{off}$} & \colhead{Count rate} & \colhead{Reduced rate} & 
1033: \colhead{Pile-up} & \colhead{Alternate names} & \colhead{Ref} \\ 
1034: \colhead{(1)} & \colhead{(2)} & \colhead{(3)} & \colhead{(4)} & \colhead{(5)} & 
1035: \colhead{(6)} & \colhead{(7)} & 
1036: \colhead{(8)} & \colhead{(9)} & \colhead{(10)} & \colhead{(11)} & \colhead{(12)} 
1037: & \colhead{(13)} & \colhead{(14)} \\
1038: }
1039: \startdata
1040: \multicolumn{14}{c}{ULXs}\\
1041: \tableline
1042: 
1043: 1 & U5  & X031820.0$-$662911 & NGC 1313    & 39.47 & 2950 & 2002 Oct 13 & 7(1/1) & 2.4 & 0.25 & 0.13 & 0.15 &                   &               \\
1044: 2 & -   & X081929.0$+$704219 & Holmberg II & 39.91 & 1564 & 2001 Nov 02 & 7(1/4) & 0.6 & 0.51 & 0.18 & 0.21 & IXO 31, ULX1, X-1 & 1, 2, 3       \\
1045: 3 & U10 & X095550.0$+$694046 & M82         & 39.52 & 1302 & 1999 Sep 20 & 3(1/1) & 0.4 & 0.12 & 0.12 & 0.14 &                   &               \\
1046: 4 & U10 & X095550.0$+$694046 &             & 39.51 & 361  & 1999 Sep 20 & 3(1/1) & 0.4 & 0.11 & 0.11 & 0.13 &                   &               \\
1047: 5 & U10 & X095550.0$+$694046 &             & 40.11 & 379  & 2000 Mar 11 & 3(1/1) & 4.2 & 0.46 & 0.09 & 0.11 &                   &               \\
1048: 6 & U36 & X132938.6$+$582505 & NGC 5204    & 39.85 & 2028 & 2001 Jan 09 & 7(1/8) & 0.6 & 0.43 & 0.10 & 0.12 &                   &               \\
1049: 7 & -   & X140319.6$-$412258 & NGC 5408    & 39.82 & 2885 & 2002 May 07 & 7(1/4) & 0.7 & 0.32 & 0.08 & 0.11 & NGC 5408 X-1      & 4, 5, 6       \\
1050: 8 & U43 & X141312.2$-$652014 & Circinus    & 39.30 & 356  & 2000 Mar 14 & 7(1/1) & 0.4 & 0.12 & 0.12 & 0.15 &                   &               \\
1051: 
1052: \tableline                                                                                                 
1053: \multicolumn{14}{c}{Comparison}\\  
1054: \tableline
1055: 
1056: 9 & -   & X095533.0$+$690033 & M81         & 38.39 & 735  & 2000 May 07 & 7(1/1) & 1.0 & 0.18 & 0.59 & 0.21 & MF97              & 1, 4, 5, 6, 7 \\
1057: 
1058: \enddata
1059: 
1060: \tablecomments{                                                    
1061: (1) Observation number;                                                 
1062: (2) Sample source name if the same as one in Table 2;
1063: (3) X-ray positions (J2000);
1064: (4) Host galaxy;
1065: (5) Approximate observed luminosity in units of erg~s$^{-1}$, in the energy band 0.3$-$8.0~keV derived 
1066: from the count rate of the reprocessed data; we assumed 
1067: a PL model with $\Gamma=$~1.8 and Galactic absorption column;
1068: (6) Observation ID;
1069: (7) Date of observation start;
1070: (8) CCD number where the object is located and subarray values (in parantheses);
1071: (9) Off axis angle of the source in arcminutes;
1072: (10) Count rate in s$^{-1}$;
1073: (11) Reduced count rate calculated for pileup estimations explained in Section~2.4.
1074: This takes into account the off-axis angle in col. (9) and the subarray values in col. (8);
1075: (12) Pile-up estimation based on the reduced count rate in column 11;
1076: (13) Common names from the literature in col. (14) (see Table~2 for common names and references for objects listed in col. 2);
1077: (14) References.
1078: }
1079: \tablerefs{
1080: 1. \citet{liub05};
1081: 2. \citet{col02}; 
1082: 3. \citet{goad06};
1083: 4. \citet{fen05};
1084: 5. \citet{lium05};
1085: 6. \citet{swa04};
1086: 7. \citet{swa03};
1087: }
1088: \end{deluxetable}
1089: 
1090: 
1091: 
1092: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Table 2 %%%%%%%%%%%%%%%%%   
1093: 
1094: 
1095: \clearpage
1096: \pagestyle{empty} 
1097: 
1098: 
1099: \begin{deluxetable}{rcccccrrccccclcl}
1100: \tablecolumns{15}                                    
1101: \rotate    
1102: \tablewidth{0pt} 
1103: \tabletypesize{\tiny}
1104: \setlength{\tabcolsep}{0.05in}
1105: \tablenum{2} %\\               
1106: \tablecaption{Properties of sample objects\label{table2}}          
1107: \tablehead{
1108: \colhead{No} & \colhead{Name} & \colhead{Position} & \colhead{Galaxy} & 
1109: \colhead{Dist} & \colhead{N$^{Gal}_H$} & \colhead{OBS} & 
1110: \colhead{CCD} & \colhead{Log L$_X$} & \colhead{Date} & \colhead{Exp} & 
1111: \colhead{Counts} & \colhead{$\theta_{off}$} & 
1112: \colhead{Alternate names} & \colhead{Loc} & \colhead{Ref} \\ 
1113: \colhead{(1)} & \colhead{(2)} & \colhead{(3)} & \colhead{(4)} & \colhead{(5)} & 
1114: \colhead{(6)} & \colhead{(7)} & 
1115: \colhead{(8)} & \colhead{(9)} & \colhead{(10)} & \colhead{(11)} & \colhead{(12)} & 
1116: \colhead{(13)} & \colhead{(14)} & \colhead{(15)} & \colhead{(16)}
1117: }
1118: \startdata
1119: \multicolumn{14}{c}{\small {ULX sample}}\\
1120: \tableline
1121: 
1122: 1  & U1   & X012435.2$+$034731 & NGC 520    & 29.6 & 3.3    & 2924 & 7(1/1) & 40.1 & 2003 Jan 29 & 49.3 & 1037.2   $\pm$ 32.3  & 2.1  & Source 11 	               & DB  & 1			  \\
1123: 2  & U2   & X013350.9$+$303938 & M33        & 0.7  & 5.69   & 787  & 7(1/4) & 39.0 & 2000 Jan 11 & 9.3  & 26733.4  $\pm$ 163.6 & 8.9  & M33 X-8	                       & N   & 2, 3, 4, 5		  \\
1124: 3  &      & X013350.9$+$303939 &            &      & 5.69   & 2023 & 7(1/1) & 38.9 & 2001 Jul 06 & 88.8 & 171301.0 $\pm$ 415.2 & 12.5 &			               &     &  			  \\
1125: 5  & U3   & X022231.4$+$422024 & NGC 891    & 9.6  & 8.12   & 794  & 7(1/1) & 39.4 & 2000 Nov 01 & 50.9 & 1977.4   $\pm$ 44.5  & 1.7  & NGC 891 X-4 	               & DB  & 6, 7			  \\
1126: 6  & U4   & X024238.9$-$000055 & M77        & 15.2 & 3.54   & 344  & 7(1/1) & 39.7 & 2000 Feb 21 & 47.4 & 1524.8   $\pm$ 39.6  & 0.7  & -		               & A   & 8			  \\
1127: 7  & U5   & X031820.0$-$662911 & NGC 1313   & 3.7  & 3.96   & 3550 & 2(1/1) & 40.1 & 2002 Nov 09 & 14.6 & 10486.7  $\pm$ 102.7 & 6.0  & IXO 7, XMM1, NGC 1313 X-1      & B   & 9, 64, 10, 11		  \\
1128: 8  & U6   & X034555.7$+$680455 & IC 342     & 3.9  & 29.39  & 2916 & 7(1/8) & 39.5 & 2002 Apr 29 & 9.3  & 2033.6   $\pm$ 45.1  & 0.5  & IXO 22, IC 342 X-7, XMM1, X-1  & A   & 9, 12, 64, 6, 13, 26, 47   \\
1129: 9  &      & X034555.6$+$680456 &            &      & 29.39  & 2917 & 7(1/8) & 39.5 & 2002 Aug 26 & 9.9  & 2191.8   $\pm$ 46.8  & 0.6  &				       &     &  			  \\
1130: 10 & U7   & X073625.5$+$653540 & NGC 2403   & 4.2  & 4.17   & 2014 & 7(1/1) & 39.2 & 2001 Apr 17 & 35.6 & 5364.2   $\pm$ 73.3  & 2.7  & Source 21, NGC 2403 X-1, XMM1  & A   & 14, 6, 64		  \\
1131: 11 & U8   & X085333.7$+$511930 & NGC 2681   & 13.3 & 2.48   & 2061 & 7(1/1) & 39.2 & 2001 May 02 & 79.0 & 1105.9   $\pm$ 33.3  & 1.3  & NGC 2681 PSX-3	   	       & D   & 15			  \\
1132: 12 & U9   & X095546.5$+$694040 & M82        & 5.2  & 4.02   & 361  & 3(1/1) & 39.0 & 1999 Sep 20 & 33.3 & 1174.2   $\pm$ 34.9  & 0.8  & Source 9	               & SF  & 16			  \\
1133: 13 & U10  & X095550.1$+$694048 &            &      & 4.03   & 378  & 3(1/1) & 40.0 & 1999 Dec 30 & 4.1  & 1404.7   $\pm$ 38.0  & 4.0  & Source 7, M82 X-1	       & SF  & 16, 17, 29		  \\
1134: 14 & U11  & X095551.0$+$694045 &            &      & 4.03   & 2933 & 7(1/1) & 39.2 & 2002 Jun 18 & 18.0 & 1595.2   $\pm$ 41.3  & 0.6  & Source 5	               & SF  & 16			  \\
1135: 15 & U12  & X095551.1$+$694043 &            &      & 4.03   & 361  & 3(1/1) & 39.1 & 1999 Sep 20 & 33.3 & 1353.6   $\pm$ 39.1  & 0.4  & Source 4	               & SF  & 16			  \\
1136: 16 & U13  & X103843.3$+$533102 & NGC 3310   & 18.7 & 1.12   & 2939 & 7(1/2) & 39.7 & 2003 Jan 25 & 47.2 & 1003.9   $\pm$ 31.7  & 0.3  & IXO 38, NGC 3310 ULX2, X-3     & A   & 9, 18, 19		  \\
1137: 17 & U14  & X103845.9$+$533012 &            &      & 1.11   & 2939 & 7(1/2) & 39.8 & 2003 Jan 25 & 47.2 & 1541.8   $\pm$ 41.0  & 0.6  & NGC 3310 X-1, X1	       & N   & 6, 18			  \\
1138: 18 & U15  & X103846.0$+$533004 &            &      & 1.11   & 2939 & 7(1/2) & 39.8 & 2003 Jan 25 & 47.2 & 1221.6   $\pm$ 35.7  & 0.7  & -			       & SF  & -			  \\
1139: 19 & U16  & X111126.0$+$554017 & M108       & 14.1 & 0.78   & 2025 & 7(1/1) & 39.4 & 2001 Sep 08 & 59.4 & 1278.9   $\pm$ 35.9  & 2.8  & Source 26		       & D   & 20			  \\
1140: 20 & U17  & X112015.8$+$133514 & NGC 3628   & 7.7  & 2.22   & 2039 & 7(1/1) & 39.3 & 2000 Dec 02 & 58.0 & 2995.8   $\pm$ 54.8  & 0.9  & IXO 39		               & DB  & 9, 21			  \\
1141: 21 & U18  & X120151.4$-$185225 & NGC 4038/9 & 21.7 & 3.95   & 3040 & 7(1/1) & 39.7 & 2001 Dec 29 & 69.0 & 1009.7   $\pm$ 31.9  & 0.9  & Source 11		       & AM  & 22, 23			  \\
1142: 22 &      & X120151.3$-$185225 &            &      & 3.95   & 3043 & 7(1/1) & 39.8 & 2002 Apr 18 & 67.1 & 1377.4   $\pm$ 37.3  & 1.0  &				       &     &  			  \\
1143: 23 &      & X120151.3$-$185225 &            &      & 3.95   & 3041 & 7(1/1) & 39.8 & 2002 Nov 22 & 72.9 & 1491.4   $\pm$ 38.8  & 0.9  &				       &     &  			  \\
1144: 24 & U19  & X120152.1$-$185134 &            &      & 3.95   & 315  & 7(1/1) & 39.9 & 1999 Dec 01 & 72.2 & 1984.1   $\pm$ 44.6  & 1.6  & Source 16		       & AM  & 22, 23			  \\
1145: 25 &      & X120152.1$-$185133 &            &      & 3.95   & 3040 & 7(1/1) & 39.8 & 2001 Dec 29 & 69.0 & 1587.0   $\pm$ 40.0  & 0.9  & 		               &     &  			  \\
1146: 26 &      & X120152.1$-$185133 &            &      & 3.95   & 3042 & 7(1/1) & 39.8 & 2002 May 31 & 67.3 & 1474.9   $\pm$ 38.5  & 1.6  &				       &     &  			  \\
1147: 27 &      & X120152.1$-$185133 &            &      & 3.95   & 3041 & 7(1/1) & 39.8 & 2002 Nov 22 & 72.9 & 1491.8   $\pm$ 38.8  & 0.8  &				       &     &  			  \\
1148: 28 & U20  & X120155.6$-$185215 &            &      & 3.96   & 315  & 7(1/1) & 39.8 & 1999 Dec 01 & 72.2 & 1344.0   $\pm$ 37.3  & 1.9  & Source 42		       & AM  & 22, 23			  \\
1149: 29 & U21  & X120156.4$-$185158 &            &      & 3.96   & 315  & 7(1/1) & 39.7 & 1999 Dec 01 & 72.2 & 1307.1   $\pm$ 36.2  & 1.6  & Source 44		       & AM  & 22, 23			  \\
1150: 30 &      & X120156.5$-$185157 &            &      & 3.96   & 3040 & 7(1/1) & 39.7 & 2001 Dec 29 & 69.0 & 1264.8   $\pm$ 35.6  & 0.4  &				       &     &  			  \\
1151: 31 & U22  & X120807.5$+$651028 & NGC 4125   & 18.1 & 1.82   & 2071 & 7(1/1) & 39.5 & 2001 Sep 09 & 64.2 & 1051.7   $\pm$ 33.0  & 0.4  & -			       & E   & 68			  \\
1152: 32 & U23  & X123030.6$+$414142 & NGC 4485   & 9.3  & 1.78   & 1579 & 7(1/1) & 39.6 & 2000 Nov 03 & 19.5 & 1450.1   $\pm$ 38.1  & 2.6  & IXO 62, NGC 4485 X-1           & A   & 9, 6, 24 		  \\
1153: 33 & U24  & X123049.2$+$122604 & M87        & 17.1 & 2.54   & 2707 & 7(1/1) & 39.3 & 2002 Jul 06 & 98.7 & 1064.6   $\pm$ 36.8  & 3.1  & -			       & E   & 68			  \\
1154: 34 & U25  & X123551.7$+$275604 & NGC 4559   & 9.7  & 0.82   & 2026 & 7(1/4) & 39.9 & 2001 Jan 14 & 9.4  & 1434.4   $\pm$ 37.9  & 0.6  & IXO 65, NGC 4559 X-1, X7       & D   & 9, 6, 25, 26, 48 	  \\
1155: 35 &      & X123551.7$+$275604 &            &      & 0.82   & 2027 & 7(1/4) & 40.1 & 2001 Jun 04 & 10.7 & 2093.2   $\pm$ 45.8  & 0.6  &				       &     &  			  \\
1156: 36 & U26  & X123558.6$+$275742 &            &      & 0.8    & 2027 & 7(1/4) & 39.8 & 2001 Jun 04 & 10.7 & 1300.9   $\pm$ 36.1  & 2.9  & IXO 66, NGC 4559 X-4, X10      & DB  & 9, 6, 25, 26, 48 	  \\
1157: 37 & U27  & X123617.4$+$255856 & NGC 4565   & 16.4 & 1.31   & 3950 & 7(1/1) & 39.8 & 2003 Feb 08 & 59.2 & 2146.5   $\pm$ 46.5  & 2.0  & IXO 67, NGC 4565 ULX4          &  B  & 9, 27, 28		  \\
1158: 38 & U28  & X123740.3$+$114728 & NGC 4579   & 20.3 & 2.52   & 807  & 7(1/4) & 40.1 & 2000 May 02 & 33.9 & 1654.6   $\pm$ 40.7  & 1.3  & NGC 4579 X-1                   & D   & 30			  \\
1159: 39 & U29  & X124155.6$+$323217 & NGC 4631   & 6.9  & 1.29   & 797  & 7(1/1) & 39.2 & 2000 Apr 16 & 59.2 & 3223.1   $\pm$ 56.8  & 0.5  & IXO 68, NGC 4631 X-1, XMM1     & SF  & 9, 6, 64 		  \\
1160: 40 & U30  & X125053.3$+$410714 & M94        & 4.3  & 1.44   & 808  & 7(1/4) & 39.0 & 2000 May 13 & 47.4 & 4472.6   $\pm$ 70.8  & 0.7  & NGC 4736 X-1		       & DB  & 30			  \\
1161: 41 & U31  & X130521.9$-$492827 & NGC 4945   & 5.2  & 14.94  & 864  & 7(1/1) & 39.1 & 2000 Jan 27 & 49.1 & 2983.5   $\pm$ 54.9  & 1.3  & NGC 4945 XMM4		       & DB  & 64			  \\
1162: 42 & U32  & X130532.9$-$492734 &            &      & 14.84  & 864  & 7(1/1) & 39.1 & 2000 Jan 27 & 49.1 & 2797.7   $\pm$ 53.2  & 0.7  & NGC 4945 X-2, XMM1             & DB  & 31, 64			  \\
1163: 43 & U33  & X131519.5$+$420302 & NGC 5055   & 7.2  & 1.3    & 2197 & 7(1/1) & 39.8 & 2001 Aug 27 & 28.0 & 2354.6   $\pm$ 48.6  & 6.0  & IXO 74, NGC 5055 X-2           & D   & 9, 6			  \\
1164: 44 & U34  & X132507.4$-$430410 & Cen A      & 4.9  & 8.41   & 316  & 3(1/1) & 39.0 & 1999 Dec 05 & 35.7 & 1108.8   $\pm$ 33.9  & 9.2  & IXO 75			       & E   & 9, 32			  \\
1165: 45 &      & X132507.5$-$430410 &            &      & 8.41   & 962  & 1(1/1) & 39.3 & 2000 May 17 & 36.5 & 2556.1   $\pm$ 50.7  & 5.5  &				       &     &  			  \\
1166: 46 & U35  & X132519.8$-$430317 &            &      & 8.4    & 316  & 3(1/1) & 39.2 & 1999 Dec 05 & 35.7 & 2124.0   $\pm$ 46.4  & 7.1  & IXO 76			       & E   & 9, 32, 63		  \\
1167: 47 & U36  & X132938.6$+$582506 & NGC 5204   & 4.8  & 1.38   & 2029 & 7(1/8) & 39.4 & 2001 May 02 & 9.0  & 1498.1   $\pm$ 38.7  & 0.6  & IXO 77, NGC 5204 X-1, XMM1     & SF  & 9, (6, 18, 26, 33, 34), 64 \\
1168: 48 & U37  & X133719.8$-$295349 & M83        & 4.7  & 3.69   & 793  & 6(1/1) & 39.0 & 2000 Apr 29 & 51.0 & 2419.2   $\pm$ 49.2  & 2.7  & IXO 82, H30, XMM1	       & D   & 9,  35, 64		  \\
1169: 49 & U38  & X140304.0$+$542735 & M101       & 5.4  & 1.15   & 4731 & 6(1/1) & 39.2 & 2004 Jan 19 & 56.2 & 3213.5   $\pm$ 56.8  & 4.4  & MF37, ULX2, H19, XMM-1, XMM2   & A   & 39, 18, 37, 40, 64	  \\
1170: 50 & U39  & X140314.3$+$541806 &            &      & 1.15   & 5309 & 7(1/1) & 39.0 & 2004 Mar 14 & 70.8 & 3889.1   $\pm$ 62.6  & 5.2  & H25, P51, XMM-2, XMM1          & A   & 37, 38, 40, 64		  \\
1171: 51 &      & X140314.3$+$541806 &            &      & 1.15   & 4732 & 7(1/1) & 39.0 & 2004 Mar 19 & 69.8 & 3902.9   $\pm$ 62.8  & 5.2  &				       &     &  			  \\
1172: 52 & U40  & X140332.4$+$542103 &            &      & 1.15   & 934  & 7(1/1) & 39.2 & 2000 Mar 26 & 98.2 & 9024.5   $\pm$ 95.1  & 3.8  & M101 X5, H32, P98, ULX-1       & A   & 18, 37, 38, 49, 65, 66, 60 \\
1173: 53 & U41  & X140414.3$+$542604 &            &      & 1.15   & 934  & 3(1/1) & 39.1 & 2000 Mar 26 & 98.2 & 3549.6   $\pm$ 59.7  & 10.6 & IXO 83, ULX3, H45, XMM-3       & D   & 9, 18, 37, 40		  \\
1174: 54 &      & X140414.1$+$542603 &            &      & 1.15   & 4731 & 2(1/1) & 38.8 & 2004 Jan 19 & 56.2 & 1085.5   $\pm$ 33.1  & 8.7  &				       &     &  			  \\
1175: 55 &      & X140414.2$+$542603 &            &      & 1.15   & 5300 & 3(1/1) & 39.2 & 2004 Mar 07 & 52.1 & 2306.5   $\pm$ 48.2  & 11.0 &				       &     &  			  \\
1176: 56 &      & X140414.2$+$542603 &            &      & 1.15   & 5309 & 3(1/1) & 38.9 & 2004 Mar 14 & 70.8 & 1592.2   $\pm$ 40.2  & 11.5 &				       &     &  			  \\
1177: 57 &      & X140414.2$+$542603 &            &      & 1.15   & 4732 & 3(1/1) & 38.8 & 2004 Mar 19 & 69.8 & 1147.7   $\pm$ 34.3  & 11.5 &				       &     &  			  \\
1178: 58 & U42  & X141310.1$-$652045 & Circinus   & 3.7  & 59.7   & 356  & 7(1/1) & 39.1 & 2000 Mar 14 & 23.1 & 1715.3   $\pm$ 41.4  & 0.9  & CG X-2, source F	       & D   & 41, 42			  \\
1179: 59 & U43  & X141312.2$-$652014 &            &      & 59.92  & 365  & 7(1/8) & 39.7 & 2000 Mar 14 & 5.0  & 1634.5   $\pm$ 40.5  & 0.4  & CG X-1, source J	       & D   & 41, 42, 43		  \\
1180: 60 & U44  & X145358.9$+$033217 & NGC 5775   & 22.4 & 3.51   & 2940 & 7(1/1) & 39.9 & 2002 Apr 05 & 58.2 & 1324.2   $\pm$ 36.4  & 1.1  & -			       & D   &  			  \\
1181: 61 & U45  & X203500.7$+$601131 & NGC 6946   & 5.5  & 20.23  & 1043 & 7(1/1) & 39.6 & 2001 Sep 07 & 58.3 & 8451.9   $\pm$ 92.1  & 4.8  & MF16, NGC 6946 X-11, 58, X8    & D   & 44, 6, 45, 46		  \\
1182: 62 &      & X203500.8$+$601131 &            &      & 20.23  & 4404 & 7(1/1) & 39.5 & 2002 Nov 25 & 30.0 & 3750.0   $\pm$ 61.3  & 2.9  &				       &     &  			  \\
1183: 63 & U46  & X225724.7$-$410344 & NGC 7424   & 11.5 & 1.33   & 3496 & 7(1/1) & 39.7 & 2002 Jun 11 & 23.9 & 1370.8   $\pm$ 37.0  & 2.2  & ULX2			       & A   & 67			  \\
1184: 64 & U47  & X225728.9$-$410212 &            &      & 1.32   & 3496 & 7(1/1) & 39.7 & 2002 Jun 11 & 23.9 & 1331.9   $\pm$ 36.5  & 0.5  & ULX1			       & D   & 67			  \\
1185:                                                      	        			
1186: 			   
1187: \tableline                              
1188: \multicolumn{14}{c}{\small {Comparison sample}}\\ 
1189: \tableline                              
1190:                                                                                  
1191:                                                                                                                                                        
1192: 1  & C1   & X001528.9$-$391319 & NGC 55     & 1.3  & 1.74  & 2255 & 0(1/1) & 38.43 & 2001 Sep 11 & 59.4 & 8553.3   $\pm$ 92.5  & 3.7  & Source 7, 6, N55	      & D   & 50, 7, 51, 52		 \\
1193: 2  & C2   & X004238.5$+$411604 & M31        & 0.7  & 6.66  & 1585 & 0(1/1) & 38.30 & 2001 Nov 19 & 4.9  & 1806.5   $\pm$ 42.6  & 4.3  & r2-26, source 35	      & DB  & (53, 62), (54, 55, 56)	 \\
1194: 3  &      & X004238.5$+$411604 &            &      & 6.66  & 2895 & 0(1/1) & 38.39 & 2001 Dec 07 & 4.9  & 2130.6   $\pm$ 46.3  & 5.3  &				      &     &				 \\
1195: 4  &      & X004238.6$+$411603 &            &      & 6.66  & 2896 & 1(1/1) & 38.40 & 2002 Feb 06 & 4.9  & 2302.4   $\pm$ 48.1  & 5.2  &				      &     &				 \\
1196: 5  &      & X004238.6$+$411604 &            &      & 6.66  & 2898 & 3(1/1) & 38.48 & 2002 Jun 02 & 4.9  & 2384.3   $\pm$ 49.0  & 6.8  &				      &     &				 \\
1197: 6  & C3   & X004305.7$+$411703 &            &      & 6.74  & 1575 & 7(1/1) & 38.27 & 2001 Oct 05 & 37.7 & 20560.8  $\pm$ 143.5 & 4.8  & -			      & DB  & 62, 55			 \\
1198: 7  & C4   & X004722.6$-$252051 & NGC 253    & 3.0  & 1.35  & 790  & 6(1/1) & 38.36 & 1999 Dec 27 & 43.5 & 1022.2   $\pm$ 32.3  & 7.8  & NGC 253 PSX-5, X21, XMM2      & A   & 15, (57, 58, 61), 64	 \\
1199: 8  & C5   & X004733.0$-$251749 &            &      & 1.37  & 969  & 7(1/1) & 38.67 & 1999 Dec 16 & 14.0 & 1150.2   $\pm$ 34.0  & 0.3  & NGC 253 PSX-2, X33, XMM1      &     & 15, (57, 58), 64  	 \\
1200: 9  &      & X004733.0$-$251749 &            &      & 1.37  & 790  & 6(1/1) & 38.85 & 1999 Dec 27 & 43.5 & 3246.7   $\pm$ 57.4  & 5.5  &				      &     &				 \\
1201: 10 & C6   & X004735.2$-$251512 &            &      & 1.39  & 790  & 6(1/1) & 38.57 & 1999 Dec 27 & 43.5 & 1811.8   $\pm$ 42.6  & 3.8  & NGC253  PSX-7, X36, XMM3      & A   & 15, (57, 58), 64  	 \\
1202: 11 & C7   & X073655.6$+$653541 & NGC 2403   & 4.2  & 4.17  & 2014 & 7(1/1) & 38.94 & 2001 Apr 17 & 35.6 & 2608.9   $\pm$ 51.1  & 1.1  & Source 20, XMM3		      & DB  & 14, 64			 \\
1203: 12 & C8   & X073702.4$+$653935 &            &      & 4.18  & 2014 & 7(1/1) & 38.72 & 2001 Apr 17 & 35.6 & 1600.1   $\pm$ 40.1  & 5.0  & Source 1, NGC 2403 X-4, XMM4  & A   & 14, 6, 64 		 \\
1204: 13 & C9   & X122809.3$+$440508 & NGC 4449   & 3.0  & 1.5   & 2031 & 7(1/1) & 38.38 & 2001 Feb 04 & 26.6 & 1138.8   $\pm$ 33.8  & 2.3  & NGC 4449 X-1, source 10       & SF  & 6, 59			 \\
1205: 14 & C10  & X122817.8$+$440634 &            &      & 1.49  & 2031 & 7(1/1) & 38.46 & 2001 Feb 04 & 26.6 & 1356.9   $\pm$ 36.9  & 1.7  & NGC 4449 X-7, source 27	      & SF  & 6, 59			 \\
1206: 15 & C11  & X124211.1$+$323236 & NGC 4631   & 6.9  & 1.29  & 797  & 7(1/1) & 38.73 & 2000 Apr 16 & 59.2 & 1104.0   $\pm$ 33.3  & 3.1  & NGC 4631 PSX-1, XMM5	      & D   & 15, 64			 \\
1207: 16 & C12  & X125050.3$+$410712 & M94        & 4.3  & 1.44  & 808  & 7(1/4) & 38.51 & 2000 May 13 & 47.4 & 1349.6   $\pm$ 37.0  & 0.3  & M94 X-4			      & DB  & 30			 \\
1208: 17 & C13  & X125052.7$+$410719 &            &      & 1.44  & 808  & 7(1/4) & 38.79 & 2000 May 13 & 47.4 & 2553.3   $\pm$ 51.6  & 0.6  & M94 X-3			      & DB  & 30			 \\
1209: 18 & C14  & X125053.1$+$410712 &            &      & 1.44  & 808  & 7(1/4) & 38.83 & 2000 May 13 & 47.4 & 2782.9   $\pm$ 53.6  & 0.6  & M94 X-2			      & DB  & 30			 \\
1210: 19 & C15  & X130518.5$-$492824 & NGC 4945   & 5.2  & 14.96 & 864  & 7(1/1) & 38.69 & 2000 Jan 27 & 49.1 & 1115.3   $\pm$ 33.8  & 1.8  & NGC 4945 XMM3		      & DB  & 64			 \\
1211: 20 & C16  & X130538.1$-$492545 &            &      & 14.74 & 864  & 6(1/1) & 38.94 & 2000 Jan 27 & 49.1 & 1535.9   $\pm$ 39.3  & 2.6  & Source 3, NGC 4945 XMM2	      & D   & 31, 64			 \\
1212: 21 & C17  & X133659.5$-$294959 & M83        & 4.7  & 3.69  & 793  & 7(1/1) & 38.54 & 2000 Apr 29 & 51.0 & 1249.9   $\pm$ 35.5  & 3.9  & H17, source 28, M83 XMM2      & A   & 35, 36, 64		 \\
1213: 22 & C18  & X133700.9$-$295203 &            &      & 3.7   & 793  & 7(1/1) & 38.58 & 2000 Apr 29 & 51.0 & 1341.9   $\pm$ 40.8  & 2.0  & source 44		      & DB  & 36			 \\
1214: 23 & C19  & X133704.3$-$295404 &            &      & 3.72  & 793  & 7(1/1) & 38.59 & 2000 Apr 29 & 51.0 & 1381.1   $\pm$ 37.2  & 1.0  & H26, source 62		      & A   & 35, 36			 \\
1215: 24 & C20  & X133704.4$-$295122 &            &      & 3.69  & 793  & 7(1/1) & 38.63 & 2000 Apr 29 & 51.0 & 1527.7   $\pm$ 39.2  & 2.2  & H27, source 64, M83 XMM3      & DB  & 35, 36, 64		 \\
1216: 25 & C21  & X140228.3$+$541627 & M101       & 5.4  & 1.14  & 5322 & 6(1/1) & 38.76 & 2004 May 03 & 64.7 & 1235.2   $\pm$ 35.3  & 5.6  & M101 XMM4		      & D   & 64			 \\
1217: 26 & C22  & X203500.1$+$600908 & NGC 6946   & 5.5  & 20.13 & 1043 & 7(1/1) & 38.91 & 2001 Sep 07 & 58.3 & 1894.7   $\pm$ 43.6  & 3.4  & IXO 85, NGC 6946 X-9, 56, X7  & A   & 9, 6, 45, 46		 \\
1218: \enddata
1219: 
1220: \tablecomments{                                                    
1221: (1) Observation number;                                                 
1222: (2) Source name;
1223: (3) X-ray positions (J2000);
1224: (4) Host galaxy;
1225: (5) Galaxy distance from Tully(1988) in Mpc;
1226: (6) Galactic absorption column in units of 10$^{20}$cm$^{-2}$;
1227: (7) Observation ID;
1228: (8) CCD number where the object is located and subarray values (in parantheses); 
1229: the subarray value represents the fraction of the CCD actually used in the observation;
1230: (9) Approximate observed luminosity in units of erg s$^{-1}$, in the energy band 0.3$-$8.0~keV derived 
1231: from the count rate of the reprocessed data; we assumed 
1232: a PL model with $\Gamma=$~1.8 and Galactic absorption column;
1233: (10) Date of observation start;
1234: (11) Exposure time in ks;
1235: (12) Net counts in the 0.3$-$8.0 keV energy band;
1236: (13) Off axis angle of the source in arcminutes; the values listed here 
1237: and the subarray values in col. (8) were used when we rejected the piledup sources;
1238: (14) Common names from the literature in column 16 (the names correspond to 
1239: references in the same order; references that use the same name are in parantheses; some papers do not give special names or the names 
1240: are given using the coordinates, these were not used);
1241: (15) Location in the galaxy; abbreviations are: A - spiral arm, D - disk, DB - disk or bulge, E - elliptical galaxy, no special location, 
1242: SF - star forming region, AM - arm in merger, N -  nucleus;
1243: (16) References
1244: }
1245: \tablerefs{
1246: 1. \citet{read05};
1247: 2. \citet{dub02};
1248: 3. \citet{col99};
1249: 4. \citet{fos04};
1250: 5. \citet{lapa03};
1251: 6. \citet{rob2000};
1252: 7. \citet{read97};
1253: 8. \citet{smi03};
1254: 9. \citet{col02};
1255: 10. \citet{col95};
1256: 11. \citet{mill04};
1257: 12. \citet{kong03};
1258: 13. \citet{sug01};
1259: 14. \citet{sch03};
1260: 15. \citet{hum03};
1261: 16. \citet{mat01};
1262: 17. \citet{stro03};
1263: 18. \citet{liub05};
1264: 19. \citet{jen04a};
1265: 20. \citet{wang03};
1266: 21. \citet{str01};
1267: 22. \citet{fab01};
1268: 23. \citet{zez02};
1269: 24. \citet{rob02};
1270: 25. \citet{vog97};
1271: 26. \citet{rob04};
1272: 27. \citet{fos04};
1273: 28. \citet{wu02}; 
1274: 29. \citet{muc06};
1275: 30. \citet{era02};
1276: 31. \citet{gua00};
1277: 32. \citet{kra01};
1278: 33. \citet{rob01};
1279: 34. \citet{rob05};
1280: 35. \citet{imm99};
1281: 36. \citet{sor03};
1282: 37. \citet{wang99};
1283: 38. \citet{pen01};
1284: 39. \citet{mat97};
1285: 40. \citet{jen04b};
1286: 41. \citet{bau01};
1287: 42. \citet{smi01};
1288: 43. \citet{wei04};
1289: 44. \citet{roco03};
1290: 45. \citet{hol03};
1291: 46. \citet{lira00};
1292: 47. \citet{rob03};
1293: 48. \citet{cro04}; 
1294: 49. \citet{muk03};
1295: 50. \citet{sch97};
1296: 51. \citet{rob97};
1297: 52. \citet{sto04};
1298: 53. \citet{kong02};
1299: 54. \citet{pri93};
1300: 55. \citet{kaa02};
1301: 56. \citet{bar03};
1302: 57. \citet{vog99};
1303: 58. \citet{pie01};
1304: 59. \citet{sum03};
1305: 60. \citet{kun05};
1306: 61. \citet{tan05};
1307: 62. \citet{wil04};
1308: 63. \citet{gho06};
1309: 64. \citet{win06};
1310: 65. \citet{muk05};
1311: 66. \citet{kong05};
1312: 67. \citet{sor06};
1313: 68. \citet{swa04};
1314: }
1315: \end{deluxetable}
1316: 
1317: 
1318: 
1319: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%                                                                               
1320: %%%%%%%%%%%%%%%%%%%%%%%%%%%% TABLE 3 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%                                                                               
1321: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%   
1322: 
1323: \begin{deluxetable}{l|ccccc|ccccc}
1324: \tablecolumns{10} 
1325: \tabletypesize{\scriptsize}                                                                 
1326: \tablewidth{0pt}                                                            
1327: \setlength{\tabcolsep}{0.03in}                                              
1328: \tablenum{3}                                                                                                                                            
1329: \tablecaption{Single-component spectral fits\label{table3}}          
1330: \tablehead{
1331: \colhead{} & \multicolumn{5}{c}{PL model$^a$} & \multicolumn{5}{c}{MCD model$^a$ } \\
1332: \tableline
1333: \colhead{Source} & 
1334: \colhead{$\Gamma^b$} & \colhead{N$_H^f$} & \colhead{Norm} & \colhead{Good} & 
1335: \colhead{$\chi^2$/} & 
1336: \colhead{kT$_{in}^c$} & \colhead{N$_H^f$} & \colhead{Norm} & \colhead{Good} & 
1337: \colhead{$\chi^2$/} \\
1338: \colhead{} & 
1339: \colhead{} & \colhead{} & \colhead{PL$^d$} & \colhead{fits$^h$} & 
1340: \colhead{d.o.f.$^g$} & 
1341: \colhead{} & \colhead{} & \colhead{MCD$^e$} & \colhead{fits$^h$} & 
1342: \colhead{d.o.f.$^g$} \\
1343: \colhead{} & \colhead{} &
1344: \colhead{(10$^{21}$~cm$^{-2}$)} & \colhead{} & \colhead{} & \colhead{(keV)} & 
1345: \colhead{(10$^{21}$~cm$^{-2}$)} & \colhead{} & \colhead{} \\
1346: \colhead{(1)} & \colhead{(2)} & \colhead{(3)} & \colhead{(4)} & \colhead{(5)} & 
1347: \colhead{(6)} & \colhead{(7)} & 
1348: \colhead{(8)} & \colhead{(9)} & \colhead{(10)} &
1349: \colhead{(11)}
1350: }
1351: \startdata
1352: \multicolumn{11}{c}{\small {ULX sample}}\\
1353: \tableline  
1354: 
1355: U1  &    2.56$^{+ 0.23}_{- 0.21}$ &     4.8$^{+  0.8}_{-  0.7}$ &  7.8$^{+12.0}_{ -1.4}\times$10$^{-5}$ &   &      68.6/53 &    0.91$^{+ 0.11}_{- 0.10}$ &     2.0$^{+  0.5}_{-  0.4}$ &  1.1$^{+ 0.8}_{ -0.4}\times$10$^{-2}$ &   &      95.5/53 \\
1356: U2  &               2.11          &                2.4          &  5.5                                  &   &   2296.5/728 &               1.07          &                      0.7    &  6.4                                  &   &   2792.4/728 \\
1357: U3  &    1.94$^{+ 0.14}_{- 0.13}$ &     7.6$^{+  0.9}_{-  0.8}$ &  1.4$^{+ 1.0}_{ -0.2}\times$10$^{-4}$ &   &     70.0/100 &    1.43$^{+ 0.14}_{- 0.12}$ &                 4.8$\pm$0.5 &  5.9$^{+ 2.4}_{ -1.7}\times$10$^{-3}$ & G &     91.8/100 \\
1358: U4  &    0.81$^{+ 0.14}_{- 0.13}$ &     5.3$^{+  1.2}_{-  1.0}$ &  3.8$^{+10.3}_{ -0.6}\times$10$^{-5}$ & G &      93.1/86 &                  ($>$ 3.61) &     5.6$^{+  0.8}_{-  0.7}$ &  1.9$^{+ 2.7}_{ -0.1}\times$10$^{-4}$ & G &      94.7/86 \\
1359: U5  &               1.70$\pm$0.05 &                 4.5$\pm$0.3 &  2.2$^{+ 0.2}_{ -0.1}\times$10$^{-3}$ & G &    304.7/289 &    1.66$^{+ 0.07}_{- 0.06}$ &                 2.5$\pm$0.2 &  6.9$^{+ 1.1}_{ -0.9}\times$10$^{-2}$ &   &    353.6/289 \\
1360: U6  &               1.71$\pm$0.09 &                 3.7$\pm$0.5 &  6.0$^{+ 0.7}_{ -0.6}\times$10$^{-4}$ & G &    215.7/215 &    1.71$^{+ 0.13}_{- 0.12}$ &                 1.3$\pm$0.3 &  1.6$^{+ 0.5}_{ -0.4}\times$10$^{-2}$ &   &    259.8/215 \\
1361: U7  &               2.17$\pm$0.07 &                 4.2$\pm$0.3 &  4.4$^{+ 0.8}_{ -0.3}\times$10$^{-4}$ &   &    260.3/182 &    1.13$^{+ 0.05}_{- 0.04}$ &                 2.1$\pm$0.2 &  3.9$^{+ 0.7}_{ -0.6}\times$10$^{-2}$ & G &    172.6/182 \\
1362: U8  &    1.88$^{+ 0.16}_{- 0.15}$ &                 1.7$\pm$0.4 &  2.3$^{+ 8.1}_{ -0.3}\times$10$^{-5}$ & G &      60.8/58 &    1.21$^{+ 0.13}_{- 0.12}$ &                 0.3$\pm$0.2 &  2.1$^{+ 0.9}_{ -0.7}\times$10$^{-3}$ & G &      53.4/58 \\
1363: U9  &    2.45$^{+ 0.22}_{- 0.20}$ &     7.0$^{+  1.3}_{-  1.2}$ &  2.0$^{+ 1.6}_{ -0.4}\times$10$^{-4}$ & G &      63.7/61 &    1.11$^{+ 0.12}_{- 0.11}$ &                 3.0$\pm$0.8 &  1.3$^{+ 0.7}_{ -0.5}\times$10$^{-2}$ &   &      76.7/61 \\
1364: U10 &    1.00$^{+ 0.21}_{- 0.20}$ &    10.5$^{+  2.7}_{-  2.4}$ &  7.3$^{+ 3.8}_{ -1.9}\times$10$^{-4}$ & G &      72.0/81 &                     $>$3.42 &     9.3$^{+  1.2}_{-  1.1}$ &                                $<$6.8 & G &      77.4/81 \\
1365: U11 &    1.16$^{+ 0.25}_{- 0.23}$ &    30.3$^{+  5.1}_{-  4.4}$ &  3.2$^{+ 3.0}_{ -1.0}\times$10$^{-4}$ & G &      82.8/96 &                    $>$ 3.23 &    26.6$^{+  2.3}_{-  2.1}$ &  9.6$^{+ 0.7}_{ -0.5}\times$10$^{-4}$ & G &      87.9/96 \\
1366: U12 &    2.81$^{+ 0.71}_{- 0.63}$ &   225.7$^{+ 45.3}_{- 38.8}$ &  7.7$^{+21.1}_{ -5.3}\times$10$^{-3}$ & G &      99.8/84 &    1.89$^{+ 0.58}_{- 0.37}$ &   179.2$^{+ 30.5}_{- 26.6}$ &  2.1$^{+ 4.2}_{ -1.5}\times$10$^{-2}$ &   &     101.8/84 \\
1367: U13 &    2.52$^{+ 0.27}_{- 0.23}$ &     2.3$^{+  0.5}_{-  0.4}$ &  4.4$^{+10.8}_{ -0.6}\times$10$^{-5}$ &   &      69.9/50 &    0.81$^{+ 0.12}_{- 0.11}$ &     0.3$^{+  0.3}_{-  0.2}$ &  1.1$^{+ 0.9}_{ -0.5}\times$10$^{-2}$ &   &     100.8/50 \\
1368: U14 &    1.46$^{+ 0.16}_{- 0.14}$ &     5.1$^{+  1.0}_{-  0.9}$ &  6.6$^{+10.2}_{ -1.0}\times$10$^{-5}$ & G &      80.9/88 &    2.07$^{+ 0.34}_{- 0.26}$ &     3.3$^{+  0.6}_{-  0.5}$ &  1.2$^{+ 0.8}_{ -0.5}\times$10$^{-3}$ & G &      87.2/88 \\
1369: U15 &    1.78$^{+ 0.18}_{- 0.16}$ &     6.5$^{+  1.3}_{-  1.1}$ &  7.3$^{+12.9}_{ -1.4}\times$10$^{-5}$ & G &      76.8/70 &    1.53$^{+ 0.19}_{- 0.16}$ &     4.2$^{+  0.8}_{-  0.7}$ &  2.8$^{+ 1.5}_{ -1.0}\times$10$^{-3}$ & G &      66.7/70 \\
1370: U16 &    1.90$^{+ 0.16}_{- 0.15}$ &     3.9$^{+  0.6}_{-  0.5}$ &  5.0$^{+ 8.7}_{ -0.7}\times$10$^{-5}$ & G &      73.1/65 &    1.30$^{+ 0.14}_{- 0.12}$ &     2.1$^{+  0.4}_{-  0.3}$ &  3.2$^{+ 1.4}_{ -1.0}\times$10$^{-3}$ & G &      65.9/65 \\
1371: U17 &               1.71$\pm$0.11 &                 7.6$\pm$0.7 &  1.5$^{+ 0.9}_{ -0.2}\times$10$^{-4}$ & G &    149.7/149 &    1.69$^{+ 0.15}_{- 0.14}$ &     5.2$^{+  0.5}_{-  0.4}$ &  4.4$^{+ 1.6}_{ -1.2}\times$10$^{-3}$ &   &    182.6/149 \\
1372: U18 &               1.79$\pm$0.08 &     4.0$^{+  0.3}_{-  0.4}$ &          4.7$\pm$0.4$\times$10$^{-5}$ &   &    270.2/208 &    1.48$^{+ 0.10}_{- 0.09}$ &                 2.1$\pm$0.2 &  2.1$^{+ 0.5}_{ -0.4}\times$10$^{-3}$ &   &    269.4/208 \\
1373: U19 &    1.15$^{+ 0.05}_{- 0.02}$ &                 0.6$\pm$0.1 &  2.2$^{+ 0.2}_{ -0.1}\times$10$^{-5}$ & G &    391.6/345 &    2.65$^{+ 0.21}_{- 0.22}$ &                     $<$ 0.1 &  2.6$^{+ 0.8}_{ -0.5}\times$10$^{-4}$ &   &    416.7/345 \\
1374: U20 &    1.22$^{+ 0.13}_{- 0.11}$ &                0.2($<$ 0.5) &  1.8$^{+ 6.8}_{ -0.2}\times$10$^{-5}$ & G &      62.2/75 &    1.89$^{+ 0.26}_{- 0.24}$ &                     $<$ 0.1 &  6.1$^{+ 4.0}_{ -2.0}\times$10$^{-4}$ & G &      84.6/75 \\
1375: U21 &    1.97$^{+ 0.11}_{- 0.10}$ &     1.4$^{+  0.3}_{-  0.1}$ &          2.8$\pm$0.2$\times$10$^{-5}$ &   &    203.7/133 &               1.16$\pm$0.08 &                     $<$ 0.2 &  2.8$^{+ 0.8}_{ -0.6}\times$10$^{-3}$ &   &    211.1/133 \\
1376: U22 &    2.00$^{+ 0.19}_{- 0.17}$ &                 0.7$\pm$0.3 &  2.1$^{+ 7.9}_{ -0.3}\times$10$^{-5}$ & G &      48.5/53 &    0.79$^{+ 0.09}_{- 0.08}$ &                     $<$ 0.1 &  9.0$^{+ 4.3}_{ -2.9}\times$10$^{-3}$ &   &      84.3/53 \\
1377: U23 &    1.80$^{+ 0.14}_{- 0.13}$ &                 3.4$\pm$0.5 &  1.5$^{+ 1.0}_{ -0.2}\times$10$^{-4}$ &   &      97.8/75 &    1.41$^{+ 0.14}_{- 0.12}$ &     1.7$^{+  0.4}_{-  0.3}$ &  8.3$^{+ 3.3}_{ -2.4}\times$10$^{-3}$ & G &      80.5/75 \\
1378: U24 &    2.52$^{+ 0.28}_{- 0.25}$ &     1.9$^{+  0.5}_{-  0.4}$ &  2.1$^{+ 9.6}_{ -0.3}\times$10$^{-5}$ & G &      59.1/66 &    0.73$^{+ 0.10}_{- 0.09}$ &                 0.3$\pm$0.2 &  8.3$^{+ 5.8}_{ -3.2}\times$10$^{-3}$ & G &      70.4/66 \\
1379: U25 &    2.25$^{+ 0.11}_{- 0.10}$ &                 1.6$\pm$0.2 &          2.5$\pm$0.2$\times$10$^{-4}$ &   &    252.3/169 &                        0.91 &                         0.1 &                                   5.0 &   &    361.5/169 \\
1380: U26 &    1.89$^{+ 0.14}_{- 0.13}$ &                 2.5$\pm$0.4 &  2.2$^{+ 1.1}_{ -0.3}\times$10$^{-4}$ & G &      70.1/67 &    1.29$^{+ 0.13}_{- 0.11}$ &     1.0$^{+  0.3}_{-  0.2}$ &  1.6$^{+ 0.6}_{ -0.4}\times$10$^{-2}$ & G &      56.8/67 \\
1381: U27 &               2.00$\pm$0.12 &                 3.5$\pm$0.4 &  8.3$^{+ 7.0}_{ -0.9}\times$10$^{-5}$ & G &    119.4/108 &    1.20$^{+ 0.10}_{- 0.09}$ &                 1.7$\pm$0.2 &  6.8$^{+ 2.3}_{ -1.7}\times$10$^{-3}$ &   &    134.6/108 \\
1382: U28 &               1.88$\pm$0.12 &                 1.7$\pm$0.3 &  7.8$^{+ 7.3}_{ -0.8}\times$10$^{-5}$ & G &      82.8/83 &    1.33$^{+ 0.12}_{- 0.11}$ &                 0.3$\pm$0.2 &  5.1$^{+ 1.9}_{ -1.4}\times$10$^{-3}$ & G &      79.4/83 \\
1383: U29 &               1.90$\pm$0.09 &     2.7$^{+  0.3}_{-  0.2}$ &  1.0$^{+ 0.6}_{ -0.1}\times$10$^{-4}$ & G &    158.7/135 &    1.28$^{+ 0.09}_{- 0.08}$ &     1.1$^{+  0.2}_{-  0.1}$ &  7.3$^{+ 2.0}_{ -1.6}\times$10$^{-3}$ &   &    204.2/135 \\
1384: U30 &    1.26$^{+ 0.07}_{- 0.06}$ &                 0.2$\pm$0.1 &  8.6$^{+ 4.2}_{ -0.5}\times$10$^{-5}$ & G &    166.7/184 &    1.86$^{+ 0.17}_{- 0.12}$ &                     $<$ 0.1 &  3.0$^{+ 0.8}_{ -0.7}\times$10$^{-3}$ &   &    265.5/184 \\
1385: U31 &    2.27$^{+ 0.11}_{- 0.10}$ &     5.7$^{+  0.6}_{-  0.5}$ &  2.5$^{+ 1.0}_{ -0.3}\times$10$^{-4}$ &   &    188.8/142 &               1.13$\pm$0.06 &     2.9$^{+  0.4}_{-  0.3}$ &  1.9$^{+ 0.5}_{ -0.4}\times$10$^{-2}$ & G &    154.9/142 \\
1386: U32 &               1.79$\pm$0.11 &     6.4$^{+  0.8}_{-  0.7}$ &  1.8$^{+ 1.0}_{ -0.2}\times$10$^{-4}$ & G &    158.9/142 &    1.60$^{+ 0.13}_{- 0.12}$ &     3.9$^{+  0.5}_{-  0.4}$ &  5.6$^{+ 1.8}_{ -1.4}\times$10$^{-3}$ & G &    153.1/142 \\
1387: U33 &    2.50$^{+ 0.14}_{- 0.13}$ &                 2.6$\pm$0.3 &  4.7$^{+ 1.1}_{ -0.5}\times$10$^{-4}$ &   &    131.5/104 &               0.80$\pm$0.06 &                 0.7$\pm$0.2 &  1.3$^{+ 0.5}_{ -0.3}\times$10$^{-1}$ &   &    179.1/104 \\
1388: U34 &                        3.97 &                         2.9 &                                   1.7 &   &    514.9/148 &                        0.26 &                         1.5 &                                   6.3 &   &    611.5/148 \\
1389: U35 &    2.42$^{+ 0.14}_{- 0.13}$ &                 2.6$\pm$0.5 &  2.3$^{+ 1.0}_{ -0.3}\times$10$^{-4}$ & G &     93.9/103 &               0.95$\pm$0.06 &                0.2($<$ 0.5) &  3.3$^{+ 1.0}_{ -0.8}\times$10$^{-2}$ & G &     84.2/103 \\
1390: U36 &    3.11$^{+ 0.22}_{- 0.20}$ &                 2.0$\pm$0.3 &  3.3$^{+ 1.2}_{ -0.3}\times$10$^{-4}$ & G &      63.8/70 &               0.49$\pm$0.05 &     0.3$^{+  0.2}_{-  0.1}$ &  5.6$^{+ 3.1}_{ -1.9}\times$10$^{-1}$ &   &     105.2/70 \\
1391: U37 &               2.43$\pm$0.12 &     1.9$^{+  0.4}_{-  0.3}$ &  1.3$^{+ 0.8}_{ -0.1}\times$10$^{-4}$ &   &    150.6/106 &               0.90$\pm$0.05 &                     $<$ 0.1 &  2.4$^{+ 0.5}_{ -0.4}\times$10$^{-2}$ &   &    154.7/106 \\
1392: U38 &               1.86$\pm$0.09 &                 3.5$\pm$0.4 &  1.5$^{+ 0.7}_{ -0.1}\times$10$^{-4}$ & G &    178.8/151 &    1.46$^{+ 0.09}_{- 0.08}$ &     1.5$^{+  0.3}_{-  0.2}$ &  6.8$^{+ 1.7}_{ -1.3}\times$10$^{-3}$ & G &    149.1/151 \\
1393: U39 &    2.17$^{+ 0.08}_{- 0.07}$ &                 1.9$\pm$0.1 &          1.0$\pm$0.1$\times$10$^{-4}$ &   &    430.4/282 &                        0.91 &                  0.5 &                                          2.1 &   &    800.3/282 \\
1394: U40 &    6.51$^{+ 0.27}_{- 0.22}$ &                 3.9$\pm$0.3 &  1.9$^{+ 0.4}_{ -0.1}\times$10$^{-4}$ &   &     315.6/84 &               0.16$\pm$0.01 &                 1.3$\pm$0.1 &  1.2$^{+ 0.4}_{ -0.3}\times$10$^{ 2}$ &   &     257.2/84 \\
1395: U41 &    3.49$^{+ 0.04}_{- 0.09}$ &     4.2$^{+  0.1}_{-  0.2}$ &          1.2$\pm$0.1$\times$10$^{-4}$ & G &    466.7/399 &               0.57$\pm$0.02 &                 0.9$\pm$0.1 &  7.4$^{+ 1.4}_{ -1.1}\times$10$^{-2}$ &   &    576.8/399 \\
1396: U42 &    1.48$^{+ 0.17}_{- 0.16}$ &     4.7$^{+  1.6}_{-  1.4}$ &  2.1$^{+ 1.6}_{ -0.4}\times$10$^{-4}$ &   &     133.0/93 &    2.25$^{+ 0.40}_{- 0.31}$ &     2.0$^{+  1.0}_{-  0.9}$ &  2.8$^{+ 1.9}_{ -1.2}\times$10$^{-3}$ &   &     140.7/93 \\
1397: U43 &    1.46$^{+ 0.17}_{- 0.16}$ &     5.3$^{+  1.7}_{-  1.5}$ &  1.0$^{+ 0.3}_{ -0.2}\times$10$^{-3}$ & G &      76.8/87 &    2.36$^{+ 0.42}_{- 0.31}$ &     2.4$^{+  1.1}_{-  1.0}$ &  1.2$^{+ 0.7}_{ -0.5}\times$10$^{-2}$ & G &      76.2/87 \\
1398: U44 &    1.90$^{+ 0.26}_{- 0.24}$ &    29.6$^{+  4.4}_{-  3.9}$ &  2.0$^{+ 2.4}_{ -0.6}\times$10$^{-4}$ & G &      84.9/75 &    1.96$^{+ 0.32}_{- 0.26}$ &    22.0$^{+  2.7}_{-  2.4}$ &  2.3$^{+ 1.7}_{ -1.0}\times$10$^{-3}$ & G &      82.9/75 \\
1399: U45 &                        2.60 &                         1.3 &                                   3.9 &   &    729.9/321 &                        0.61 &                         0.0 &                                   3.3 &   &   1436.3/321 \\
1400: U46 &    2.29$^{+ 0.18}_{- 0.16}$ &                 2.1$\pm$0.4 &  1.1$^{+ 1.0}_{ -0.1}\times$10$^{-4}$ & G &      67.4/66 &               0.92$\pm$0.10 &                 0.3$\pm$0.2 &  2.0$^{+ 1.0}_{ -0.7}\times$10$^{-2}$ &   &      99.4/66 \\
1401: U47 &               1.80$\pm$0.14 &                 2.3$\pm$0.4 & 10.0$^{+ 8.5}_{ -1.2}\times$10$^{-5}$ & G &      62.5/70 &    1.29$^{+ 0.14}_{- 0.12}$ &                 0.9$\pm$0.2 &  7.4$^{+ 3.3}_{ -2.3}\times$10$^{-3}$ & G &      69.9/70 \\
1402: 
1403: 
1404: \tableline          
1405: \multicolumn{11}{c}{\small Comparison sample} \\                                                   
1406: \tableline 
1407: 
1408: C1  &    3.69$^{+ 0.10}_{- 0.09}$ &                 4.7$\pm$0.2 & 10.0$^{+ 1.1}_{ -0.6}\times$10$^{-4}$ &   &    377.3/157 &               0.55$\pm$0.02 &                 1.3$\pm$0.1 &  6.4$^{+ 1.3}_{ -1.0}\times$10$^{-1}$ &   &    561.5/157 \\
1409: C2  &               1.49$\pm$0.05 &     1.3$^{+  0.3}_{-  0.2}$ &  8.1$^{+ 0.6}_{ -0.5}\times$10$^{-4}$ &   &    590.0/447 &               1.86$\pm$0.08 &                     $<$ 0.1 &          2.2$\pm$0.3$\times$10$^{-2}$ &   &    560.6/447 \\
1410: C3  &               2.85$\pm$0.04 &                 2.5$\pm$0.1 &  1.5$^{+ 0.1}_{ -0.0}\times$10$^{-3}$ &   &    530.1/229 &               0.72$\pm$0.01 &                         0.3 &  5.5$^{+ 0.5}_{ -0.4}\times$10$^{-1}$ &   &    442.3/229 \\
1411: C4  &    2.63$^{+ 0.28}_{- 0.24}$ &                 2.0$\pm$0.7 &  7.4$^{+14.0}_{ -1.3}\times$10$^{-5}$ &   &      66.9/51 &               0.74$\pm$0.07 &                     $<$ 0.2 &  2.6$^{+ 1.2}_{ -0.8}\times$10$^{-2}$ &   &      88.7/51 \\
1412: C5  &    1.97$^{+ 0.09}_{- 0.08}$ &                 3.8$\pm$0.3 &          2.3$\pm$0.2$\times$10$^{-4}$ &   &    249.4/204 &               1.34$\pm$0.08 &                 1.7$\pm$0.2 &  1.2$^{+ 0.3}_{ -0.2}\times$10$^{-2}$ &   &    322.1/204 \\
1413: C6  &               2.27$\pm$0.14 &     5.3$^{+  0.7}_{-  0.6}$ &  1.7$^{+ 1.0}_{ -0.2}\times$10$^{-4}$ & G &     102.9/90 &    1.09$^{+ 0.08}_{- 0.07}$ &                 2.7$\pm$0.4 &  1.5$^{+ 0.5}_{ -0.4}\times$10$^{-2}$ & G &      95.5/90 \\
1414: C7  &               1.79$\pm$0.10 &     2.7$^{+  0.4}_{-  0.3}$ &  1.4$^{+ 0.7}_{ -0.1}\times$10$^{-4}$ & G &    139.4/126 &    1.40$^{+ 0.11}_{- 0.10}$ &                 1.1$\pm$0.2 &  7.9$^{+ 2.4}_{ -1.9}\times$10$^{-3}$ & G &    141.3/126 \\
1415: C8  &    1.43$^{+ 0.12}_{- 0.11}$ &     1.5$^{+  0.4}_{-  0.3}$ &  6.1$^{+ 7.3}_{ -0.7}\times$10$^{-5}$ & G &      80.1/82 &    1.88$^{+ 0.25}_{- 0.20}$ &                 0.5$\pm$0.2 &  1.7$^{+ 0.8}_{ -0.6}\times$10$^{-3}$ & G &      79.9/82 \\
1416: C9  &    2.71$^{+ 0.24}_{- 0.21}$ &                 1.9$\pm$0.4 &  8.3$^{+ 9.9}_{ -1.1}\times$10$^{-5}$ &   &      73.2/54 &               0.62$\pm$0.06 &                 0.3$\pm$0.2 &  6.0$^{+ 3.0}_{ -1.9}\times$10$^{-2}$ &   &      68.2/54 \\
1417: C10 &    1.97$^{+ 0.17}_{- 0.16}$ &     6.8$^{+  0.9}_{-  0.8}$ &  1.7$^{+ 1.2}_{ -0.3}\times$10$^{-4}$ &   &      81.8/68 &    1.38$^{+ 0.17}_{- 0.15}$ &     4.1$^{+  0.6}_{-  0.5}$ &  7.7$^{+ 4.1}_{ -2.7}\times$10$^{-3}$ &   &     103.5/68 \\
1418: C11 &    2.69$^{+ 0.34}_{- 0.31}$ &    42.9$^{+  6.3}_{-  5.5}$ &  5.2$^{+ 4.9}_{ -1.9}\times$10$^{-4}$ & G &      46.5/62 &    1.37$^{+ 0.18}_{- 0.15}$ &    29.5$^{+  3.7}_{-  3.3}$ &  8.6$^{+ 6.4}_{ -3.7}\times$10$^{-3}$ & G &      44.3/62 \\
1419: C12 &    1.54$^{+ 0.14}_{- 0.13}$ &                 0.6$\pm$0.3 &  3.0$^{+ 7.5}_{ -0.3}\times$10$^{-5}$ &   &      85.3/70 &    1.40$^{+ 0.15}_{- 0.14}$ &                     $<$ 0.1 &  2.2$^{+ 0.9}_{ -0.6}\times$10$^{-3}$ &   &     100.9/70 \\
1420: C13 &    1.81$^{+ 0.10}_{- 0.09}$ &                 0.6$\pm$0.2 &  6.2$^{+ 5.6}_{ -0.5}\times$10$^{-5}$ & G &     96.5/119 &               1.08$\pm$0.07 &                     $<$ 0.1 &  1.0$^{+ 0.3}_{ -0.2}\times$10$^{-2}$ & G &    140.9/119 \\
1421: C14 &    2.28$^{+ 0.13}_{- 0.12}$ &                 1.0$\pm$0.2 &  7.9$^{+ 6.9}_{ -0.7}\times$10$^{-5}$ &   &    147.6/108 &               0.65$\pm$0.04 &                     $<$ 0.1 &  6.1$^{+ 1.5}_{ -1.3}\times$10$^{-2}$ &   &    233.6/108 \\
1422: C15 &    1.46$^{+ 0.15}_{- 0.14}$ &     1.6$^{+  0.6}_{-  0.5}$ &  3.7$^{+ 9.6}_{ -0.5}\times$10$^{-5}$ & G &      54.2/62 &    1.77$^{+ 0.28}_{- 0.21}$ &     0.5$^{+  0.4}_{-  0.3}$ &  1.2$^{+ 0.7}_{ -0.5}\times$10$^{-3}$ & G &      48.2/62 \\
1423: C16 &    1.86$^{+ 0.15}_{- 0.14}$ &                 4.7$\pm$0.9 &  1.0$^{+ 1.1}_{ -0.2}\times$10$^{-4}$ & G &      84.5/78 &    1.51$^{+ 0.16}_{- 0.14}$ &                 2.1$\pm$0.6 &  3.8$^{+ 1.7}_{ -1.2}\times$10$^{-3}$ & G &      85.6/78 \\
1424: C17 &    1.38$^{+ 0.15}_{- 0.14}$ &     0.7$^{+  0.4}_{-  0.3}$ &  2.7$^{+ 7.3}_{ -0.3}\times$10$^{-5}$ & G &      74.2/66 &    1.81$^{+ 0.27}_{- 0.21}$ &                     $<$ 0.1 &  8.9$^{+ 5.4}_{ -3.1}\times$10$^{-4}$ &   &      83.5/66 \\
1425: C18 &    2.60$^{+ 0.22}_{- 0.20}$ &                 2.0$\pm$0.4 &  5.5$^{+ 9.4}_{ -0.7}\times$10$^{-5}$ & G &      65.8/74 &    0.74$^{+ 0.08}_{- 0.07}$ &                0.1($<$ 0.3) &  2.0$^{+ 1.1}_{ -0.7}\times$10$^{-2}$ & G &      74.2/74 \\
1426: C19 &    2.60$^{+ 0.17}_{- 0.16}$ &                 3.8$\pm$0.5 &  8.6$^{+ 9.3}_{ -1.2}\times$10$^{-5}$ & G &      74.5/67 &               0.87$\pm$0.07 &                 1.3$\pm$0.3 &  1.5$^{+ 0.6}_{ -0.4}\times$10$^{-2}$ & G &      65.4/67 \\
1427: C20 &    2.35$^{+ 0.17}_{- 0.15}$ &                 3.0$\pm$0.4 &  7.5$^{+ 8.4}_{ -0.9}\times$10$^{-5}$ & G &      79.3/75 &               0.93$\pm$0.08 &     0.9$^{+  0.3}_{-  0.2}$ &  1.3$^{+ 0.5}_{ -0.4}\times$10$^{-2}$ & G &      83.0/75 \\
1428: C21 &    2.30$^{+ 0.18}_{- 0.17}$ &     5.6$^{+  0.9}_{-  0.8}$ &  8.4$^{+11.4}_{ -1.4}\times$10$^{-5}$ &   &      85.6/63 &    1.07$^{+ 0.10}_{- 0.09}$ &     2.8$^{+  0.6}_{-  0.5}$ &  8.1$^{+ 3.4}_{ -2.4}\times$10$^{-3}$ & G &      68.8/63 \\
1429: C22 &    5.55$^{+ 0.54}_{- 0.45}$ &     5.3$^{+  0.9}_{-  0.8}$ &  3.2$^{+ 2.2}_{ -0.5}\times$10$^{-4}$ &   &     136.9/73 &    0.26$^{+ 0.03}_{- 0.02}$ &                 1.5$\pm$0.5 &                 6.2$^{+ 6.4}_{ -3.0}$ &   &     163.0/73 \\
1430: 
1431: \enddata
1432: \tablenotetext{a}{Model names}
1433: \tablenotetext{b}{Photon index for the PL model}
1434: \tablenotetext{c}{Temperature of the accretion disk at inner radius for the MCD model}
1435: \tablenotetext{d}{Normalization constant for the PL model, in units of photons 
1436: keV$^{-1}$~cm$^{-2}$~s$^{-1}$ at 1~keV.}
1437: \tablenotetext{e}{Normalization constant for the MCD model, in units of 
1438: R$_{in}$(km)$^2$~cos$\theta/$~D(10~kpc)$^2$, where
1439: R$_{in}$(km) is the inner radius of the accretion disk in units of km, 
1440: cos$\theta$ is the cosine of the inclination
1441: of the accretion disk from the line of sight, and D(10~kpc) is the distance to 
1442: the source in units of 10~kpc.}
1443: \tablenotetext{f}{Intrinsic absorbing hydrogen column density, in units of 
1444: 10$^{21}$~cm$^{-2}$} 
1445: \tablenotetext{g}{$\chi^2$ value for the fit and number of degrees of freedom}
1446: \tablenotetext{h}{The ``Good'' fits, marked with a "G", have 
1447: $\chi_{\nu}^{2}\leq~$1.2}
1448: \end{deluxetable}
1449: 
1450: 
1451: %%%%%%%%%%%%%%%%%%%%%%%%%%%% TABLE 4 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1452: 
1453: \begin{deluxetable}{l|c|cc|c|cc|c|cc|c|cc}
1454: \rotate
1455: \tablecolumns{12}
1456: \tablewidth{0pt}                                                            
1457: \tabletypesize{\tiny}   
1458: \setlength{\tabcolsep}{0.05in}                                            
1459: \tablenum{4}                                                                                                                                           
1460: \tablecaption{Statistical tests for single-component fits\label{table4}}          
1461: \tablehead{
1462: \colhead{Samples} & 
1463: \colhead{$\Gamma$ (PL) $^a$} & \colhead{K-S} & \colhead{T-test} &
1464: \colhead{kT$_{in}$ (MCD) $^b$} & \colhead{K-S} & \colhead{T-test} &
1465: \colhead{N$_H$ (PL) $^c$} & \colhead{K-S} & \colhead{T-test} &
1466: \colhead{N$_H$ (MCD) $^d$} & \colhead{K-S} & \colhead{T-test}\\
1467: \colhead{(1)} & 
1468: \colhead{(2)} & \colhead{(3)} & \colhead{(4)} & 
1469: \colhead{(5)} & \colhead{(6)} & \colhead{(7)} & 
1470: \colhead{(8)} & \colhead{(9)} & \colhead{(10)} &
1471: \colhead{(11)} & \colhead{(12)} & \colhead{(13)}
1472: }
1473: \startdata
1474: ULX-All vs. Comp-All     &  2.11$\pm$0.89 vs. 2.36$\pm$0.92 & 0.155       & 0.324       &  1.44$\pm$0.86 vs. 1.14$\pm$0.47 & 0.457       & 0.126       &  21.51$\pm$0.50 vs. 21.42$\pm$0.42 & 0.74 & 0.49 & 21.18$\pm$0.66 vs. 21.08$\pm$0.55 & 0.84 & 0.59  \\	     
1475: ULX-GF vs. Comp-GF    	 &  1.88$\pm$0.59 vs. 2.02$\pm$0.50 & 0.731       & 0.479       &  1.81$\pm$0.99 vs. 1.25$\pm$0.37 & 0.075       & 0.078       &  21.56$\pm$0.59 vs. 21.44$\pm$0.50 & 0.70 & 0.55 & 21.33$\pm$0.59 vs. 21.12$\pm$0.63 & 0.32 & 0.39  \\
1476: ULX-HL vs. ULX-LL        &  1.70$\pm$0.47 vs. 2.44$\pm$1.05 & {\bf 0.011} & {\bf 0.005} &  1.89$\pm$1.02 vs. 1.12$\pm$0.50 & {\bf 0.015} & {\bf 0.001} &  21.54$\pm$0.51 vs. 21.51$\pm$0.50 & 0.75 & 0.86 & 21.25$\pm$0.70 vs. 21.13$\pm$0.64 & 0.47 & 0.58  \\
1477: ULX-HL-GF vs. ULX-LL-GF	 &  1.52$\pm$0.39 vs. 2.18$\pm$0.57 & {\bf 0.020} & {\bf 0.001} &  2.26$\pm$1.10 vs. 1.22$\pm$0.27 & {\bf 0.022} & {\bf 0.012} &  21.58$\pm$0.60 vs. 21.54$\pm$0.61 & 0.52 & 0.87 & 21.52$\pm$0.59 vs. 21.09$\pm$0.53 & 0.43 & 0.11  \\
1478: ULX-HL vs. Comp-All      &  1.70$\pm$0.47 vs. 2.36$\pm$0.92 & {\bf 0.021} & {\bf 0.009} &  1.89$\pm$1.02 vs. 1.14$\pm$0.47 & {\bf 0.050} & {\bf 0.003} &  21.54$\pm$0.51 vs. 21.42$\pm$0.42 & 0.72 & 0.43 & 21.25$\pm$0.70 vs. 21.08$\pm$0.55 & 0.37 & 0.45  \\
1479: ULX-HL-GF vs. Comp-GF	 &  1.52$\pm$0.39 vs. 2.02$\pm$0.50 & 0.111       & {\bf 0.009} &  2.26$\pm$1.10 vs. 1.25$\pm$0.37 & {\bf 0.023} & {\bf 0.008} &  21.58$\pm$0.60 vs. 21.44$\pm$0.50 & 0.48 & 0.53 & 21.52$\pm$0.59 vs. 21.12$\pm$0.63 & 0.18 & 0.15  \\
1480: ULX-LL vs. Comp-All 	 &  2.44$\pm$1.05 vs. 2.36$\pm$0.92 & 0.916       & 0.699       &  1.12$\pm$0.50 vs. 1.14$\pm$0.47 & 0.896       & 0.897       &  21.51$\pm$0.50 vs. 21.42$\pm$0.42 & 0.62 & 0.51 & 21.13$\pm$0.64 vs. 21.08$\pm$0.55 & 0.98 & 0.81  \\
1481: ULX-LL-GF vs. Comp-GF	 &  2.18$\pm$0.50 vs. 2.02$\pm$0.50 & 0.727       & 0.460       &  1.22$\pm$0.27 vs. 1.25$\pm$0.37 & 0.637       & 0.843       &  21.54$\pm$0.61 vs. 21.44$\pm$0.50 & 0.93 & 0.64 & 21.09$\pm$0.53 vs. 21.12$\pm$0.63 & 0.97 & 0.90  \\
1482: \enddata                                                       
1483: \tablecomments{     
1484: Statistical comparison using the results from single-component fits.
1485: The samples compared in the first column are defined in section 3.1.
1486: The abreviations are: GF, good fits, with $\chi_{\nu}^{2}\leq~$1.2;
1487: HL, high luminosity, ULXs with X-ray (absorbed) luminosity L$_X\ge$~5.0$\times$10$^{39}$~erg~s$^{-1}$;
1488: LL, low luminosity, ULXs with L$_X\le$~5.0$\times$10$^{39}$~erg~s$^{-1}$.
1489: For each pair of samples in first column we performed both K-S and the T-test for the means,
1490: and we calculate the corresponding probabilities. The significant differences, 
1491: with probabilities $\leq~$0.05, are shown in bold.
1492: High luminosity ULXs have harder spectra than both low-luminosity ULXs and the comparison sample.
1493: There is also marginal evidence that ULXs show higher disk temperatures 
1494: than the comparison sample if we only consider the good fits in both samples.}
1495: \tablenotetext{a}{Average photon index in the PL model and 1 $\sigma$ errors}
1496: \tablenotetext{b}{Average inner disk temperature in keV for the MCD model and one sigma errors}
1497: \tablenotetext{c}{Average log H column density for the PL model and 1 $\sigma$ errors in units of cm$^{-2}$}
1498: \tablenotetext{d}{The same for the MCD model}
1499: \end{deluxetable}
1500: 
1501: 
1502: %%%%%%%%%%%%%%%%%%%%%%%%%%%% Table 5 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%                                                                                  
1503: 
1504: 
1505: \clearpage
1506: \pagestyle{empty}  
1507: 
1508: \begin{deluxetable}{l|cccccccc}
1509: \tablecolumns{8}   
1510: \tablewidth{0pt}                                                                                   
1511: \tabletypesize{\tiny}                                                                           
1512: \tablenum{5}                                                                                       
1513: \tablecaption{Two-component spectral fits (model PLMCD) \label{table5}}          
1514: \tablehead{
1515: \colhead{Source} & 
1516: \colhead{$\Gamma^a$} & \colhead{kT$_{in}^b$} & \colhead{Norm} & \colhead{Norm} & \colhead{N$_H^e$} & \colhead{$\Delta\chi^2$/} & \colhead{Good} & \colhead{$\chi^2$/} 
1517: \\
1518: \colhead{} & 
1519: \colhead{} & \colhead{} & \colhead{PL$^c$} & \colhead{MCD$^d$} & \colhead{} & \colhead{Prob.$^f$} & \colhead{fits$^g$} & \colhead{d.o.f.$^h$} 
1520: \\
1521: \colhead{(1)} & \colhead{(2)} & \colhead{(3)} & \colhead{(4)} & \colhead{(5)} & 
1522: \colhead{(6)} & \colhead{(7)} & 
1523: \colhead{(8)} &
1524: \colhead{(9)}
1525: }
1526: \startdata
1527: \multicolumn{8}{c}{\small {ULX sample}} \\
1528: \tableline  
1529: 
1530: U1  &    2.20$^{+ 0.46}_{- 0.69}$ &    0.28$^{+ 0.17}_{- 0.12}$ &  4.9$^{+13.2}_{ -3.2}\times$10$^{-5}$ &                 1.1$^{+ 1.0}_{ -0.9}$ &     5.0$^{+  2.6}_{-  1.4}$ &         3.1/0.69 (0.69) &   &      65.5/51 \\
1531: U2  &    2.52$^{+ 0.26}_{- 0.15}$ &    1.34$^{+ 0.03}_{- 0.06}$ &  4.0$^{+ 0.3}_{ -0.1}\times$10$^{-3}$ &                  1.4$\times$10$^{-1}$ &     2.5$^{+  0.2}_{-  0.1}$ & 397.9/$>$0.99 ($>$0.99) &   &    950.9/424 \\
1532: U13 &    2.23$^{+ 0.38}_{- 0.36}$ &    0.13$^{+ 0.04}_{- 0.02}$ &  3.7$^{+10.3}_{ -1.3}\times$10$^{-5}$ &               $<$6.8$\times$10$^{ 3}$ &     5.0$^{+  2.0}_{-  1.6}$ &  27.4/$>$0.99 ($>$0.99) & G &      42.5/48 \\
1533: U18 &    1.82$^{+ 0.30}_{- 0.17}$ &    0.19$^{+ 0.09}_{- 0.05}$ &  3.0$^{+10.2}_{ -0.5}\times$10$^{-5}$ &               $<$4.6$\times$10$^{ 2}$ &     6.7$^{+  4.1}_{-  2.5}$ &         8.0/0.94 (0.93) &   &      68.1/51 \\
1534: U19 &    1.19$^{+ 0.10}_{- 0.14}$ &    0.20$^{+ 0.02}_{- 0.19}$ &  2.1$^{+ 0.2}_{ -0.3}\times$10$^{-5}$ &                 1.1$^{+ 0.9}_{ -1.0}$ &                     $<$ 4.4 &         0.8/0.09 (0.18) &   &     100.0/74 \\
1535: U25 &    1.76$^{+ 0.28}_{- 0.30}$ &    0.19$^{+ 0.08}_{- 0.05}$ &  1.6$^{+ 1.0}_{ -0.5}\times$10$^{-4}$ &               $<$5.8$\times$10$^{ 2}$ &     2.0$^{+  1.4}_{-  0.8}$ &  12.7/$>$0.99 ($>$0.99) &   &      87.0/70 \\
1536: U25 &    2.16$^{+ 0.19}_{- 0.21}$ &    0.13$^{+ 0.03}_{- 0.02}$ &  3.0$^{+ 1.0}_{ -0.6}\times$10$^{-4}$ &               $<$9.8$\times$10$^{ 3}$ &     3.8$^{+  1.4}_{-  1.0}$ &  33.7/$>$0.99 ($>$0.99) &   &     112.3/93 \\
1537: U33 &    2.34$^{+ 0.23}_{- 0.37}$ &    0.20$^{+ 0.13}_{- 0.06}$ &  4.0$^{+ 1.4}_{ -1.6}\times$10$^{-4}$ &               $<$4.7$\times$10$^{ 2}$ &     2.7$^{+  1.0}_{-  0.7}$ &         3.4/0.73 (0.71) &   &    128.2/102 \\
1538: U34 &    3.21$^{+ 0.75}_{- 0.72}$ &    0.10$^{+ 0.02}_{- 0.01}$ &  9.6$^{+22.2}_{ -4.0}\times$10$^{-5}$ &  1.4$^{+ 0.8}_{ -1.2}\times$10$^{ 5}$ &     9.0$^{+  0.8}_{-  1.1}$ &  63.6/$>$0.99 ($>$0.99) & G &      58.0/49 \\
1539: U34 &    2.73$^{+ 0.19}_{- 0.16}$ &              0.11$\pm$ 0.01 &  2.2$^{+ 0.7}_{ -0.3}\times$10$^{-4}$ &  3.1$^{+15.0}_{ -2.4}\times$10$^{ 4}$ &     7.0$^{+  1.3}_{-  1.1}$ & 149.4/$>$0.99 ($>$0.99) &   &     130.1/93 \\
1540: U37 &    3.35$^{+ 0.43}_{- 0.70}$ &    1.14$^{+ 0.15}_{- 0.23}$ &  1.0$^{+ 2.7}_{ -0.3}\times$10$^{-4}$ &  6.1$^{+ 5.9}_{ -2.4}\times$10$^{-3}$ &     2.4$^{+  2.7}_{-  1.4}$ &        11.2/0.98 (0.96) &   &    139.4/104 \\
1541: U39 &    1.52$^{+ 0.23}_{- 0.29}$ &    0.25$^{+ 0.06}_{- 0.05}$ &         4.8$\pm$ 1.6$\times$10$^{-5}$ &                 3.7$^{+ 8.1}_{ -2.4}$ &     2.1$^{+  0.5}_{-  0.4}$ &  54.9/$>$0.99 ($>$0.99) & G &    142.8/139 \\
1542: U39 &    1.28$^{+ 0.23}_{- 0.25}$ &              0.26$\pm$ 0.04 &  3.8$^{+ 5.6}_{ -1.1}\times$10$^{-5}$ &                 3.6$^{+ 5.5}_{ -2.0}$ &               2.1$\pm$  0.4 &  83.9/$>$0.99 ($>$0.99) & G &    148.5/137 \\
1543: U40 &    3.77$^{+ 0.50}_{- 0.42}$ &              0.13$\pm$ 0.01 &  3.2$^{+ 4.9}_{ -0.9}\times$10$^{-5}$ &  4.6$^{+ 3.2}_{ -1.4}\times$10$^{ 2}$ &     1.8$^{+  0.3}_{-  0.2}$ & 169.5/$>$0.99 ($>$0.99) &   &     146.2/82 \\
1544: U42 &                        1.82 &                        0.13 &                                   3.6 &                                   5.7 &                        11.6 &         1.6/0.42 (0.47) &   &     131.4/91 \\
1545: U45 &    2.43$^{+ 0.12}_{- 0.10}$ &              0.12$\pm$ 0.01 &  3.5$^{+ 0.6}_{ -0.4}\times$10$^{-4}$ &  4.2$^{+ 8.9}_{ -2.4}\times$10$^{ 3}$ &     4.1$^{+  0.7}_{-  0.5}$ & 225.6/$>$0.99 ($>$0.99) & G &    212.6/182 \\
1546: U45 &    2.28$^{+ 0.18}_{- 0.14}$ &              0.13$\pm$ 0.02 &  2.9$^{+ 0.9}_{ -0.5}\times$10$^{-4}$ &  2.8$^{+13.5}_{ -2.0}\times$10$^{ 3}$ &     4.7$^{+  1.2}_{-  0.9}$ &  96.5/$>$0.99 ($>$0.99) &   &    160.7/133 \\
1547: 
1548: 
1549: \tableline                                                                             
1550: \multicolumn{8}{c}{\small {Comparison sample}} \\                                                 
1551: \tableline 
1552: 
1553: C1  &    3.72$^{+ 0.11}_{- 0.08}$ &              0.10$\pm$ 0.01 &         1.2$\pm$ 0.1$\times$10$^{-3}$ &  1.3$^{+ 3.3}_{ -0.9}\times$10$^{ 5}$ &               8.9$\pm$  0.4 & 137.6/$>$0.99 ($>$0.99) &   &    239.7/155 \\
1554: C2  &    1.73$^{+ 0.13}_{- 0.18}$ &    0.12$^{+ 0.02}_{- 0.01}$ &  1.1$^{+ 0.3}_{ -0.1}\times$10$^{-3}$ &  1.3$^{+12.6}_{ -1.1}\times$10$^{ 4}$ &     4.7$^{+  1.6}_{-  2.1}$ &         8.5/0.95 (0.94) &   &    144.4/108 \\
1555: C3  &    3.61$^{+ 0.66}_{- 0.38}$ &    0.82$^{+ 0.03}_{- 0.04}$ &  1.0$^{+ 0.2}_{ -0.1}\times$10$^{-3}$ &  2.1$^{+ 0.5}_{ -0.4}\times$10$^{-1}$ &     2.7$^{+  0.7}_{-  0.4}$ & 189.0/$>$0.99 ($>$0.99) &   &    341.1/227 \\
1556: C4  &    2.43$^{+ 0.42}_{- 0.60}$ &    0.18$^{+ 0.17}_{- 0.02}$ &  6.1$^{+14.2}_{ -3.5}\times$10$^{-5}$ &  1.7$^{+ 2.9}_{ -1.6}\times$10$^{ 1}$ &     3.2$^{+  4.1}_{-  2.0}$ &         3.0/0.68 (0.65) &   &      63.9/49 \\
1557: C5  &              1.96$\pm$ 0.19 &    0.20$^{+ 0.07}_{- 0.05}$ &  2.3$^{+ 1.3}_{ -0.5}\times$10$^{-4}$ &               $<$7.0$\times$10$^{ 2}$ &     6.4$^{+  2.2}_{-  1.6}$ &     12.8/$>$0.99 (0.98) & G &    164.7/139 \\
1558: C9  &               2.54($<$3.49) &    0.59$^{+ 0.06}_{- 0.14}$ &  2.8$^{+ 9.0}_{ -2.8}\times$10$^{-5}$ &  5.2$^{+11.6}_{ -1.2}\times$10$^{-2}$ &     1.1$^{+  1.4}_{-  0.9}$ &         6.8/0.92 (0.90) &   &      66.4/52 \\
1559: C10 &    1.99$^{+ 0.17}_{- 0.19}$ &              0.18($<$ 0.74) &  1.8$^{+ 0.4}_{ -0.3}\times$10$^{-4}$ &               $<$8.5$\times$10$^{ 1}$ &               9.7$\pm$  0.6 &         2.2/0.59 (0.60) &   &      79.6/66 \\
1560: C12 &    1.42$^{+ 0.23}_{- 0.51}$ &                        0.31 &  2.5$^{+ 7.5}_{ -1.5}\times$10$^{-5}$ &               $<$1.4$\times$10$^{-1}$ &                0.6($<$ 1.3) &         1.2/0.39 (0.39) &   &      84.0/68 \\
1561: C14 &    1.54$^{+ 0.37}_{- 0.52}$ &    0.35$^{+ 0.08}_{- 0.07}$ &  3.0$^{+ 6.9}_{ -1.6}\times$10$^{-5}$ &  4.3$^{+ 5.6}_{ -2.3}\times$10$^{-1}$ &     0.5$^{+  0.4}_{-  0.3}$ &  17.4/$>$0.99 ($>$0.99) &   &    130.2/106 \\
1562: C22 &    3.49$^{+ 0.62}_{- 0.56}$ &              0.12$\pm$ 0.02 &  8.9$^{+36.8}_{ -3.6}\times$10$^{-5}$ &  2.2$^{+20.7}_{ -1.7}\times$10$^{ 3}$ &     5.3$^{+  1.7}_{-  1.1}$ &  57.4/$>$0.99 ($>$0.99) & G &      79.6/71 \\
1563: 
1564: 
1565: \enddata  
1566:                                                          
1567: \tablenotetext{a}{Photon index for the PL model}
1568: \tablenotetext{b}{Temperature of the accretion disk at the inner radius for the MCD model in keV}
1569: \tablenotetext{c}{Normalization constant for the PL model as in Table~3}
1570: \tablenotetext{d}{Normalization constant for the MCD model, as in Table~3}
1571: \tablenotetext{e}{Intrinsic absorbing H column density in units of 10$^{21}$~cm$^{-2}$}
1572: \tablenotetext{f}{F-test $\Delta\chi^2$/ confidence levels for the model PLMCD against the PL model alone.
1573: The values in parentheses are obtained from simulations; see Section~3.2 for details.}
1574: \tablenotetext{g}{The ``good'' fits are marked with a ``G'' as in Table~3}
1575: \tablenotetext{h}{The $\chi^2$ value for the fit and number of degrees of freedom} 
1576: 
1577:                                                              
1578: \end{deluxetable}
1579: 
1580: 
1581: 
1582: 
1583: \clearpage
1584: \pagestyle{plaintop}
1585: 
1586: 
1587: %Fig 1
1588: \begin{figure}
1589: \rotate
1590: \epsscale{0.95}
1591: %\plottwo{luminosity.ps}{counts.ps}
1592: \plotone{f1.eps}
1593: \caption{
1594: Normalized histograms of net counts for the ULX and comparison samples.
1595: For multiple observations we use the highest number of counts for each object.
1596: The histograms are normalized to unit area.
1597: The data with counts $>$10$^5$ are from one source: the long observation of M33~X-8 (U2).
1598: }
1599: \end{figure}
1600: 
1601: 
1602: 
1603: 
1604: % fig 2
1605: \begin{figure}
1606: %\includegraphics[angle=-90,scale=0.9]{normalized.ps}
1607: \plottwo{f2a.eps}{f2b.eps}
1608: \caption{
1609: Normalized histograms from single-component fits.
1610: The histograms are normalized to have a unit area.
1611: (a) Photon index distribution from PL fits.
1612: (b) Inner disk temperature distribution from MCD fits.
1613: }
1614: \end{figure}
1615: 
1616: 
1617: 
1618: %Fig 3
1619: \begin{figure}
1620: \rotate
1621: %\plottwo{lumin-pow.ps}{lumin-mcd.ps}
1622: \plottwo{f3a.eps}{f3b.eps}
1623: \caption{
1624: (a) Luminosity vs. photon index from PL model fits. 
1625: (b) Luminosity vs. inner disk temperature using the MCD model.
1626: In the upper left corner of the left panel we define a subsample of 9 unique ULXs
1627: (U19 has 4 observations). They have luminosities $>$5$\times$10$^{39}$~erg~s$^{-1}$ 
1628: and $\Gamma<$~1.7. For clarity, we label only the objects in this ULX subsample plus any 
1629: objects with multiple observations.
1630: }
1631: \end{figure}
1632: 
1633: 
1634: %Fig 4
1635: \begin{figure}
1636: %\rotate
1637: %\plottwo{fixed-kt0.25.ps}{fixed-kt1.0.ps}
1638: \plottwo{f4a.eps}{f4b.eps}
1639: \caption{
1640: Histograms for photon indices from spectral fits with fixed inner disk temperatures,
1641: for ULX and lower luminosity samples, both normalized to unit area for easy comparison.
1642: We also show the high-luminosity, hard ULXs (filled blue regions).
1643: No significant difference is seen between ULXs and the comparison sample,
1644: but the high-luminosity ULXs are distinctly harder (i.e., flatter spectra).
1645: a) Model PLMCD0.25.
1646: b) Model PLMCD1.0.
1647: }
1648: \end{figure}
1649: 
1650: 
1651: 
1652: %Fig 5
1653: \begin{figure}
1654: \rotate
1655: %\plotone{abs_ratios-gamma.ps}
1656: \plotone{f5.eps}
1657: \caption{
1658: Ratio of MCD blackbody flux to the total flux (MCD fraction), plotted  against photon index,
1659: using the free parameter model PLMCD. The fluxes are absorbed.
1660: For clarity, we only label the 9 high-luminosity, hard ULXs as defined in Figure~3a.
1661: These have both the hardest spectra and the lowest flux contribution from the MCD components.
1662: For U11 the fraction is below 0.001. 
1663: }
1664: \end{figure}
1665: 
1666: 
1667: %Fig 6
1668: \begin{figure}
1669: \rotate
1670: %bad_lumin.ps
1671: \plottwo{f6a.eps}{f6b.eps}
1672: \caption{ 
1673: Absorbed luminosity scatter plots from the two-component spectral model with free parameters (PLMCD).
1674: We present results only for the spectra that did not provide acceptable fits with single component models.
1675: a) Photon index dependence.
1676: b) Disk temperature dependence.
1677: }
1678: \end{figure}
1679: 
1680: 
1681: 
1682: \end{document}
1683: 
1684: