------------------------------------------------------------------------ ms_gcnews.tex ApJ Letters (in press) X-MailScanner-Information: Please contact postmaster@aoc.nrao.edu for more information X-MailScanner: Found to be clean X-MailScanner-SpamCheck: not spam, SpamAssassin (score=0, required 5, autolearn=disabled) X-MailScanner-From: silas@head.cfa.harvard.edu % http://hea-www.harvard.edu/ChaMPlane/index.html % astro-ph/0509783 \documentclass{emulateapj} \newcommand{\vdag}{(v)^\dagger} \newcommand{\myemail}{silas@head.cfa.harvard.edu} %% modification to maximize figure/text placement \renewcommand{\textfraction}{.01} \renewcommand{\topfraction}{1.00} \slugcomment{To appear in ApJ Letters} \shorttitle{Galactic Center} \shortauthors{Laycock et al.} %% This is the end of the preamble. Indicate the beginning of the %% paper itself with \begin{document}. \begin{document} %% LaTeX will automatically break titles if they run longer than %% one line. However, you may use \\ to force a line break if %% you desire. \title{Constraining the Nature of the Galactic Center X-ray Source Population} %% Use \author, \affil, and the \and command to format %% author and affiliation information. %% Note that \email has replaced the old \authoremail command %% from AASTeX v4.0. You can use \email to mark an email address %% anywhere in the paper, not just in the front matter. %% As in the title, you can use \\ to force line breaks. \author{S. Laycock\altaffilmark{1}, J. Grindlay\altaffilmark{1}, M. van den Berg\altaffilmark{1}, P. Zhao\altaffilmark{1}, J. Hong\altaffilmark{1}, X. Koenig\altaffilmark{1}, E. M. Schlegel\altaffilmark{1}, S. E. Persson\altaffilmark{2}} \altaffiltext{2}{Observatories of the Carnegie Institution of Washington, 813 Santa Barbara St., Pasadena, CA, 91101} %% Notice that each of these authors has alternate affiliations, which %% are identified by the \altaffilmark after each name. Specify alternate %% affiliation information with \altaffiltext, with one command per each %% affiliation. %% Mark off your abstract in the ``abstract'' environment. In the manuscript %% style, abstract will output a Received/Accepted line after the %% title and affiliation information. No date will appear since the author %% does not have this information. The dates will be filled in by the %% editorial office after submission. \begin{abstract} We searched for infrared counterparts to the cluster of X-ray point sources discovered by Chandra in the Galactic Center Region (GCR). While the sources could be white dwarfs, neutron stars, or black holes accreting from stellar companions, their X-ray properties are consistent with magnetic Cataclysmic Variables, or High Mass X-ray Binaries (HMXB) at low accretion-rates. A direct way to decide between these possibilities and hence between alternative formation scenarios is to measure or constrain the luminosity distribution of the companions. Using infrared (J, H, K, Br${\gamma}$) imaging, we searched for counterparts corresponding to typical HMXB secondaries: spectral type B0V with K$<$15 at the GCR. We found no significant excess of bright stars in Chandra error circles, indicating that HMXBs are not the dominant X-ray source population, and account for fewer than 10\% of the hardest X-ray sources. \end{abstract} %% Keywords should appear after the \end{abstract} command. The uncommented %% example has been keyed in ApJ style. See the instructions to authors %% for the journal to which you are submitting your paper to determine %% what keyword punctuation is appropriate. \keywords{ Galaxy: center, X-rays: binaries } %% From the front matter, we move on to the body of the paper. %% In the first two sections, notice the use of the natbib \citep %% and \citet commands to identify citations. The citations are %% tied to the reference list via symbolic KEYs. The KEY corresponds %% to the KEY in the \bibitem in the reference list below. We have %% chosen the first three characters of the first author's name plus %% the last two numeral of the year of publication as our KEY for %% each reference. \section{Introduction} With the high spatial resolution of Chandra and ground-based infrared (IR) instrumentation, the Galactic Center, heavily obscured by dust, is now accessible to a wide range of astrophysical investigations. A series of Chandra observations of the GCR ($\sim$10$'$ region around SgrA*), totaling 670 ksec, have recently been analyzed by \cite{muno2003}(hereafter, M03), revealing 2357 point sources. The 2-8 keV X-ray luminosity ($10^{31-33}$ erg s$^{-1}$) and spectral index ($\Gamma < 1$) of the majority of the sources are inconsistent with normal stars, active binaries, or young stellar objects, none of which are normally persistently luminous enough at $>$2.5 keV to produce the observed fluxes in the presence of the high absorption column (N$_{H}\sim 10^{23} cm^{-2}$). The only known sources expected to be detected given the absorption are compact objects accreting from binary companions. The luminosity range of the GCR sources covers magnetic CVs and non-Roche-lobe filling HMXBs accreting in a low mass-transfer regime, e.g.~X~Per, A0535+26 \citep{negueruela2000}. Quiescent low mass X-ray binaries (LMXBs) with neutron star (NS) or black hole (BH) primaries are possible contributors but are generally too faint (BHs) or too soft (NSs). HMXBs are young objects (few My), while CVs are slow to form and long-lived. The two possibilities point to different evolutionary phases of the GCR. If HMXBs comprise a significant fraction of the GCR sources, they must be the products of recent, massive-star formation. Frequent star-burst events in the GCR are expected on $\sim$20 My scales based upon the instability of large gas clouds orbiting within 200pc of the nucleus \citep{stark2004} and the presence of several young clusters of massive stars. The HMXB hypothesis was proposed and discussed in detail by \cite{pfahl2002}. \cite{grindlay1985} suggested that the Galactic Bulge contains the debris of tidally-disrupted globular clusters. Were the GCR sources dominated by magnetic CVs, globular clusters would then contribute because CVs are over-produced per unit mass in globulars relative to the Galactic plane. \cite{morris1993} proposed a third scenario in which compact stellar remnants form in the Galactic bulge and sink toward the center of the dense mass distribution via gravitational interactions. The presence of a super-massive black hole (BH) at the center accelerates the process, leading to a central density cusp. This scenario fits with the \cite{muno2005} (M05) discovery of a centrally concentrated overabundance of X-ray transients, as well as the overall 1/$\theta$ source distribution found by M03. All three channels may be active now or have been in the past, so by measuring relative numbers of the various end-products, we hope to piece together the composition and history of the GCR. In this Letter we present preliminary results of our deep IR survey targeted at measuring the HMXB fraction. The occurrence of IR counterparts consistent with highly reddened massive stars and giants will constrain the relative abundance of HMXBs and CVs. This work is part of our Chandra Multi-wavelength Plane Survey (ChaMPlane: \citealt{grindlay2005}) which is an effort to map the space density of all types of low-luminosity accretion sources in the Galaxy. \section{Observations} The GCR is heavily obscured with average A$_{V}$$\sim$25 so counterparts can only be detected with IR imaging. The stellar density is also very high so that reaching a sufficient limiting magnitude to make meaningful constraints on population models demands excellent image quality. In June 2004 we observed the inner 10$'$x10$'$ of the GCR with the PANIC imager on the 6.5m Magellan (Baade) telescope at Las Campanas, Chile. PANIC has a 2$'$x2$'$ field of view with 0.125$''$ pixels. Under good seeing conditions (median FWHM $\sim$ 0.5$''$), we obtained a mosaic of 25 pointings in J, H, Ks, and Br$\gamma$ filters using a 5-point dither pattern with 3 exposures at each position. Sky emission was removed using off-source exposures, and the images were reduced and stacked in IRAF. Photometry was performed on the stacked images using SExtractor \citep{bertin1996}. Astrometric and photometric calibration were performed using the 2MASS point-source catalog. We used a version of X-ray point source catalog of M03, kindly provided by M. Muno, listing J2000 co-ordinates with net count rate, flux, and source significance determined in 3 energy bands S1 (0.3-1.2keV), S2 (1.2-2.5 keV), and H (2.5-8 keV). The intermediate S2 band provides some additional information, but for this Letter we restrict ourselves to discussion of GCR (H only) and foreground (S1) sources. \section{Experimental Design} \label{sect:design} We set out to identify all potential IR counterparts to the M03 X-ray sources and apply statistical tests to measure their numbers relative to random positional coincidence. Our scientific goals are to set stringent upper-limits on the number of counterparts and examine their photometric properties. The availability of X-ray spectral information and multi-band IR photometry permits a clean separation of sources situated in the GCR from those in the foreground. The GCR sources are characterized by X-ray spectra completely cut off at low energies due to absorbing gas and dust. Of the sources in the M03 catalog, 80\% are detected at $>$3$\sigma$ in the H band only, while any source detected in the S1 band must lie in the foreground. The spatial distributions of X-ray point sources and K-band stars follow similar radial profiles about the GCR. M03 demonstrated a surface density with 1/$\theta$ dependence with angular offset $\theta$ from SgrA*. Similar results have been reported for stars from IR observations (e.g. \citealt{catchpole1990}). There are $\sim$100 known Galactic HMXBs, of which 70\% have Be star companions whose spectral types lie in the narrow range O9-B2 and peak at B0V \citep{negueruela1998}. We adopt the luminosity and color of a B0V to represent ``typical'' HMXB values ($M_{V}$=$-$4, V$-$K=$-$0.83; \citealt{allen}). Note that many non-Be HMXBs are {\it more} luminous because they have super-giant companions. For the CV case, we adopt a K0V dwarf (M$_{V}$=5.9,V$-$K=1.96) at the brighter end of the normal range \citep{patterson1998}. Adopting $A_{V}$=25 and $D$=8 kpc for the GCR, our target HMXB should have an apparent magnitude K=14.4 while the assumed CV would have K=21. We note the intermediate polar GK Per with bright sub-giant counterpart would have K$\sim$19. \subsection{Photometric Completeness Curve} Due to severe crowding in the GCR, we infer the star density from our observations by correcting for incompleteness i.e. the loss of faint stars due to their brighter neighbors. Completeness curves for our PANIC photometry as functions of magnitude and distance from SgrA* were determined by simulation. For each of the 25 K-band stacked images, a random sample of stars was added using IRAF {\it mkstar} with a Gaussian PSF matched to the seeing. The photometry was repeated and the artificial stars identified and compared to their input magnitudes. A master catalog of input and recovered stars was then assembled, from which completeness curves could be extracted for any region of the 10$'$x10$'$ mosaic. The 50\% completeness limit (fraction of stars recovered within $\pm$0.5 mag of their input magnitude) within the inner 1$'$ radius occurs at K=14.5 and over the rest of the mosaic at K=15.4. Curves for star density are presented in Figure~\ref{fig:photcom} and completeness values for the full field reported in the final column of Table~\ref{tab:peakupresults}. %% FIGURE 1 \begin{figure} \includegraphics[angle=-90,width=7.5cm]{f1.ps} \caption{Star density curves for PANIC K-band photometry, in 5 magnitude ranges as indicated by legend. Observed number-density is shown by smooth lines, completeness corrected shown by histogram. A radial profile of the form A+B$\times\theta^{-1}$ is also shown. \label{fig:photcom} } \end{figure} \subsection{Peak-up Test} The IR and X-ray catalogs are repeatedly matched over a large grid of positional offsets using integer multiples of the PANIC pixel-size of 0.125$''$. At each offset position, the number of X-ray sources with one or more stars lying in their error-circle is recorded in a series of magnitude bins. We used 95\% confidence error-circles determined from the formula of \cite{hong2005} which is dependent on net counts and distance from the Chandra aimpoint. For the GCR sources, error radii are 0.3$''$-1$''$. The presence of real counterparts is signaled by a peak at the offset co-ordinates corresponding to perfect alignment of the two catalogs, providing a boresight. To confirm that X-ray and IR catalogs are on the same astrometric frame we centroided the soft-source peak, finding a residual offset of ${\Delta}$RA=0.011(1)$''$, ${\Delta}$Dec=0.092(1)$''$ with FWHM=0.55$''$. %% FIGURE 2 \begin{figure} \includegraphics[angle=-90,width=7.5cm]{f2.ps} \caption{Peak-Up test for hard and soft X-ray sources, using PANIC K-band photometry. Solid and dotted lines show projections along the RA and Dec axes of the number of X-ray sources with one or more stars inside the 95\% error circle. A prominent peak is seen at zero offset only for the soft sources. The random match level and its fluctuations are apparent beyond 1$''$ offset. \label{fig:peakup} } \end{figure} The number of random matches ($N_{ran}$) is calculated as the mean of all trials at shifts greater than 2$''$, i.e. far away from correct alignment, and the standard deviation ($\sigma$) of these values is used to characterize the uncertainty. We then take the (peak) number of matches ($N_{m}$), and subtract the random level to obtain the excess-above-random ($N_{exc}$). We claim detection of a population of counterparts above that predicted to occur by chance if $N_{exc}$ exceeds $3 \times \sigma$. The upper-limit on the total number of true counterparts was determined by monte-carlo simulation. At each iteration the X-ray catalog was offset by randomly generated shifts in X and Y such that the absolute shift was between 2$''$ and 5$''$. A randomly selected sample of $N_{true}$ X-ray sources were flagged as ``true counterparts'' and the catalogs matched as previously described. Since all of the X/IR matches were by definition random (due to the offsets) we were able to count the total number of matches $N_{sim}$ in each magnitude bin, correcting for duplicates (sources flagged as simulated matches which also happened to have a random IR counterpart) such that $N_{sim}$ = $N_{ran}$ + $N_{true}$ - $N_{ran.AND.true}$. Guided by the values of $\sigma$ from the peakup test, we ran simulations with $N_{true}$=5-40 with 100 iterations at each value. For each value of $N_{true}$ we recorded the largest value of $N_{sim}$ reached in 90\% of trials, and fitted the results with a straight line relationship. Thus the upper-limit on the number of true matches corresponds to the value of $N_{true}$ when $N_{sim}$ = $N_{m}$, these values are given in Table~\ref{tab:peakupresults} as $U_{90}$. \section{Constraints on IR/X-ray Counterparts} We performed the peak-up test first on the entire field, and then independently on a series of concentric annular regions centered on SgrA*. This allows us to limit the effects of crowding in the IR, and to follow the cusp distribution of hard X-ray sources. For the full field we found no significant excess in the number of IR matches for hard X-ray sources, irrespective of magnitude, while for soft sources we found a highly significant ($\sim$30$\sigma$) counterpart population dominated by bright (K$\sim$11) stars. This result is illustrated visually by Figure~\ref{fig:peakup} and in more detail in Table~\ref{tab:peakupresults}. There is a marginal signal ($2.9\sigma$) for hard sources versus the entire PANIC catalog, which could not be associated with any specific magnitude range. \begin{deluxetable}{lrrrrrrrr} \tabletypesize{\scriptsize} \tablecaption{Peakup Test on the full 10$'$x10$'$ field. \label{tab:peakupresults}} \tablehead{ \colhead{K$_1$-K$_2$} & \colhead{$N_m$} & \colhead{$N_{ran}$} & \colhead{$\sigma$} & \colhead{$N_{exc}$} & \colhead{Sig} & \colhead{C} & \colhead{{\it U$_{90}$}} & \colhead{{\it U$_{c}$}} } \startdata \multicolumn{3}{l}{Hard Sources (1453)} \\ $>$12 & 399 & 353 & 15.8 & 46 & 2.9 & - & 87$^{*}$ & - \\ $>$17 & 86 & 77.9 & 8.1 & 8.1 & 1.0 & - & 20$^{*}$ & - \\ 16-17 & 72 & 57.7 & 7.2 & 14.3& 2.0 & 0.3 & 54 & 10.7 \\ 15-16 & 84 & 83.0 & 9.3 & 1.0 & 0.1 & 0.5 & 27 & 7.3 \\ 14-15 & 83 & 67.5 & 8.3 & 15.5& 1.9 & 0.7 & 37 & 5.4 \\ 13-14 & 37 & 35.7 & 6.1 & 1.3 & 0.2 & 0.8 & 10 & 3.0 \\ 12-13 & 30 & 22.6 & 4.5 & 7.4 & 1.6 & 0.9 & 13 & 2.3 \\ 11-12 & 11 & 8.6 & 2.4 & 2.4 & 1.0 & 1.0 & 11 & 1.4 \\ 10-11 & 6 & 5.3 & 1.9 & 0.7 & 0.4 & 1.0 & 6 & 0.6 \\ 9-10 & 2 & 2.2 & 1.3 & -0.2& -0.2 & 1.0 & 2 & 0.2 \\ \multicolumn{3}{l}{Soft Sources (105)} \\ $>$12 & 67 & 28.6 & 4.7 & 38.4& 8.2 & - & 59$^{*}$ & - \\ 15-16 & 8 & 6.2 & 2.6 & 1.8 & 0.7 & 0.5 & 10 & 92 \\ 14-15 & 21 & 5.3 & 2.3 & 15.7& 6.8 & 0.7 & 27 & 84 \\ 13-14 & 10 & 2.6 & 1.4 & 7.4 & 5.3 & 0.8 & 11 & 59 \\ 12-13 & 13 & 1.5 & 1.3 & 11.5& 8.8 & 0.9 & 14 & 49 \\ 11-12 & 21 & 0.4 & 0.7 & 20.6& 29.4 & 1.0 & 21 & 35 \\ 10-11 & 12 & 0.2 & 0.5 & 11.8& 23.6 & 1.0 & 12 & 15 \\ 9-10 & 4 & 0.0 & 0.1 & 4.0 & 40 & 1.0 & 4 & 4 \enddata \tablecomments{{\tt In the $\sim$ 10$'$x10$'$ PANIC survey area we consider: 184$\times10^3$ K-band stars, 1453 hard X-ray sources, 105 soft X-ray sources. 2MASS point-source catalog used for K$<$12. {\em $N_m$}= Number of X-ray sources with one or more matches. {\em $N_{ran}$}= random matches, {\em $\sigma$} uncertainty in number of random matches. {\em $N_{exc}$}= number of "excess" matches above random level. {\em Sig} significance of excess. {\em C} Completeness fraction for K-band stars. {\em {\it U$_{90}$}}= Completeness corrected upper-limit on the number of counterparts at 90\% significance. {\em {\it U$_{c}$}}= Cumulative {\it U$_{90}$} brightward of K$_2$ expressed as \% of number of X-ray sources. *No completeness correction applied for unbounded mag ranges. }} \end{deluxetable} We therefore conclude the GCR counterpart population is dominated by stars fainter than our confusion limit, although some fraction of the hard matches are undoubtedly real. The 90\% confidence upper limit for the full catalog indicates there are up to 87 hard counterparts although we cannot at this stage identify them from among the 399 total matches. By placing completeness-corrected upper-limits on the number of counterparts as a function of K magnitude, we can constrain the proportion of HMXBs, and hence place a lower-limit on the population of CVs and LMXBs. The final column of Table~\ref{tab:peakupresults} gives the cumulative percentage obtained by adding the upper-limits for each magnitude range and dividing by the number of X-ray sources. We find that a maximum of 5.4\% (at 90\% confidence) of the sources could have counterparts brighter than K=15, given average GCR extinction. The presence of a significant HMXB population would appear between K=14-16, and the fact that we {\it do} detect soft-source counterparts for K=14-15 (at 6.8$\sigma$) demonstrates that we are not merely being defeated by crowding in this range. For annular bins 0-1$'$, 1$'$-2$'$, 2$'$-3$'$, 3$'$-4$'$, 4$'$-5$'$, our results constrain the HMXB population as a function of distance from SgrA*. The effect of the nuclear cluster is apparent in the inner 1$'$, where a large number of massive stars raise the upper-limits such that K$<$15 HMXB can potentially account for 9\% of the 110 hard X-ray sources present (at 90\% confidence). The limit is 5-8\% for the remaining annuli. Overall the annular results are less constraining than the full field results due to the smaller sample sizes involved. A single annular bin (2$'$-3$'$) produced a peak at 3.8$\sigma$ significance, possibly corresponding to a drop in the confusion limit sufficient to allow counterparts to be isolated. Such an effect could be due to seeing, anomalous extinction or an unrelated star cluster. We can account for 84\% of the 105 soft sources down to K$<$15, and essentially all to K$<$16 where crowding dominates. The majority of soft sources will be dMe and other foreground coronal stars. For example at spectral type M0V our 50\% completeness limit of K$\sim$15.4 implies a distance of just 1kpc, indeed 86\% of the S1 sources have counterparts from our optical ChaMPlane survey \cite{zhao2003}. \section{Photometry of Potential Counterparts} Color-magnitude diagrams (CMD) were constructed to compare the H--K vs K distribution for stars lying in X-ray error-circles with that of the field population. Extinction due to dust lying between us and the Galactic Center makes the GCR stars appear systematically redder than foreground stars, largely overcoming any degeneracy between intrinsic brightness and distance. In figure~\ref{fig:colormag} we see the soft counterparts are mainly un-reddened stars, while the hard counterparts cluster around H--K$\sim$1.6 with a large scatter. Intrinsic H--K colors do not exceed $\sim$+0.3 for any stars (\citealt{allen}). Soft-source counterparts are also brighter than average field stars, the K-S probability that they are the same is 2.6$\times 10^{-9}$. This result is consistent with the X-ray selection and peak-up test, the soft sources are shown to lie in the foreground, while the hard sources are dominated by random matches to stars in the GCR. %% FIGURE 3 \begin{figure} \includegraphics[angle=-90,width=8.5cm]{f3.ps} \caption{Color-Magnitude diagram comparing the IR counterparts of hard and soft X-ray sources with the field population. Total numbers in each class given in histogram at right. (Saturation begins at K$\lesssim$11.5) \label{fig:colormag}} \end{figure} Matches were found for 3 out of the 8 periodic X-ray sources found by \cite{muno2004} (7 in PANIC field, 5 observed in Ks, see Fig.~\ref{fig:colormag}). One of these (CXOGCR J174534.5-290201) is heavily reddened, while the others have slightly lower H--K values than the main concentration of stars in the CMD. The colors of the candidate stars do not agree with the predicted extinction from X-ray spectral fits of \cite{muno2004}. The X-ray derived N$_{H}$ of $\gtrsim$10$^{23}$cm$^{-2}$ requires the stars to be extremely bright to explain the observed K magnitudes and predicts H--K colors 2 mag redder than what is observed. These observations may be reconciled if most of the X-ray-derived N$_{H}$ arises in the binary, in an accretion column or disk obscuring the X-ray source. A column of a few 10$^{22}$cm$^{-2}$ is typical for HMXB pulsars, but is insufficient to make up the difference. It is also possible that these matches are random and the real counterparts are much fainter. No match was found for the 7 bright transients detected by M05. A preliminary search for Br$\gamma$ emission objects among X-ray IR counterparts yielded 11 candidates with S/N$>$5, of which 3 are foreground objects. Br$\gamma$--K is severely restricted by the small amplitude of the expected signal ($\sim$-0.1 mag for EW(Br$\gamma$)$\sim$10-20\AA) and the effects of crowding. \section{Conclusions} The highly absorbed GCR X-ray sources appear to be dominated by a population of accreting binaries in which the mass-donor is less massive than in a typical HMXB. This conclusion follows from the lack of a significant excess of stars brighter than K$\lesssim$15 in the error circles of hard X-ray sources, which is the apparent magnitude of a B0V star at the distance and mean Av of the GCR. By applying our peak-up test as a function of magnitude and distance from SgrA*, and correcting for completeness, we set 90\% confidence upper limits on the number of hard X-ray sources that can have counterparts in a series of ranges between K=9-16. After accumulating these limits for all counterparts brighter than K=15, such stars can account for at most 79 out of 1453 hard X-ray sources, or 5.4\%. Pushing one magnitude fainter to include Be HMXBs as late as B2, we find an upper-limit of 103 or 7.1\%. Within 1$'$ of SgrA* these upper-limits become 9\% and 15\% due to the high density of very luminous stars in the nuclear cluster. We note that large variations in A$_v$ and/or star clustering could modify these numbers. Our constraints imply non-HMXBs account for more than 90\% of the M03 X-ray sources (at least beyond 1$'$) favoring the leading alternative hypothesis of magnetic CVs. High angular resolution spectroscopy will be essential to identify the true counterparts which likely exhibit B$\gamma$ or He emission lines indicative of an accretion disk. This work was funded in part by Chandra grant AR4-5003A and NSF grant AST-0098683. We thank the referee M. Muno for his insightful suggestions. %% FIGURES %% TABLES %% If you wish to include an acknowledgments section in your paper, %% separate it off from the body of the text using the \acknowledgments %% command. %% Included in this acknowledgments section are examples of the %% AASTeX hypertext markup commands. Use \url without the optional [HREF] %% argument when you want to print the url directly in the text. Otherwise, %% use either \url or \anchor, with the HREF as the first argument and the %% text to be printed in the second. %%\acknowledgments %% The reference list follows the main body and any appendices. %% Use LaTeX's thebibliography environment to mark up your reference list. %% Note \begin{thebibliography} is followed by an empty set of %% curly braces. If you forget this, LaTeX will generate the error %% "Perhaps a missing \item?". %% %% thebibliography produces citations in the text using \bibitem-\cite %% cross-referencing. Each reference is preceded by a %% \bibitem command that defines in curly braces the KEY that corresponds %% to the KEY in the \cite commands (see the first section above). %% Make sure that you provide a unique KEY for every \bibitem or else the %% paper will not LaTeX. The square brackets should contain %% the citation text that LaTeX will insert in %% place of the \cite commands. %% We have used macros to produce journal name abbreviations. %% AASTeX provides a number of these for the more frequently-cited journals. %% See the Author Guide for a list of them. %% Note that the style of the \bibitem labels (in []) is slightly %% different from previous examples. The natbib system solves a host %% of citation expression problems, but it is necessary to clearly %% delimit the year from the author name used in the citation. %% See the natbib documentation for more details and options. %% \aap %% \aj %% \apjs %% \apj %% \apjl %% \mnras %% \pasp \begin{thebibliography}{} \bibitem[Allen(2000)]{allen} Allen's Astrophysical Quantities, 4th edition, Cox, A., (Editor), Springer-Verlag, New York. \bibitem[Bertin \& Arnouts(1996)]{bertin1996} Bertin, E., Arnouts, S., 1996, A\&AS, 117, 393 %% 1R^-1 radial profile. \bibitem[Catchpole, Whitelock \& Glass(1990)]{catchpole1990} Catchpole, R. M., Whitelock, P. A., \& Glass, I. S., 1990, MNRAS, 247, 479 %%\bibitem[Drimmel et al.(2003)]{drimmel2003} %%Drimmel, R., Cabrera-Lavers, A., Lopez-Corredoira, M., 2003, \apj, 409, 205 \bibitem[Grindlay(1985)]{grindlay1985} Grindlay, J., 1985, IAU Syp., 113, Dynamics of Star Clusters., ed. J. Goodman \& P. Hut. (Dordrecht-Kluwer), 43 %%CHAMPLANE %%\bibitem[Grindlay et al.(2003)]{grindlay2003} %%Grindlay, J., Zhao, P., Hong, J. S., Jenkins, J., Kim, Dong-Woo, Schlegel, E., Drake, J., Kashyap, V., %%Edmonds, P., Cohn, H., Lugger, P., Cool, A., 2003, AN, 324, 57 \bibitem[Grindlay et al.(2005)]{grindlay2005} Grindlay, J., et al., 2005, \apj, in press %%\bibitem[Ghenzel et al.(2003)]{ghenzel2003} %%Ghenzel R., Schodel, T., Ott, T., Eisenhauer, F., Hofmann, R., Lehnert, M., 2003, \apj, 594, 812 %%\bibitem[Heinke et al.(2003)]{heinke2003} %%Heinke, C., O., Grindlay, J., E., Lugger, P., M., Cohn, H. N., Edmonds, P. D., Lloyd, D. A., Cool, A. M., \apj, 598, 501 \bibitem[Hong et al.(2005)]{hong2005} Hong, J., et al., 2005, ApJ, in press. %% deep Chandra catalog \bibitem[Muno et al.(2003)]{muno2003} Muno, M. P., et al., 2003, \apj, 589, 225 %%Muno, M. P., Baganoff, F. K., Bautz, M. W., et al., 2003, \apj, 589, 225 %% spectra and variability \bibitem[Muno et al.(2004)]{muno2004} Muno, M. P., et al., 2004, \apj, 613, 1179 %%Muno, M. P., Arabadjis, J. S., Baganoff, F. K., et al., 2004, \apj, 613, 1179 %% transients \bibitem[Muno et al.(2005)]{muno2005} Muno, M. P., et al., 2005, \apjl, 622, 113 %%Muno, M. P., Pfahl, E., Baganoff, F. K., et al., 2005, \apjl, 622, 113 \bibitem[Morris(1993)]{morris1993} Morris, M., 1993, \apj, 408, 496 %%\bibitem[Mouawad et al.(2005)]{mouawad2005} %%Mouawad, N., Eckart, A., Pfalzner, S., Schodel, R., Moultaka, J., Spurzem, R., 2005, AN, astro-ph/0402338v3 \bibitem[Negueruela(1998)]{negueruela1998} Negueruela, I., 1998, A\&A, 338, 505 \bibitem[Negueruela et al.(2000)]{negueruela2000} Negueruela, I., et al. 2000, A\&A, 356, 1003 %%Negueruela, I., Reig, P., Finger, M. H., Roche, P., A\&A, 356, 1003 \bibitem[Patterson(1998)]{patterson1998} Patterson, J., 1998, \pasp, 110, 1132 \bibitem[Pfahl et al.(2002)]{pfahl2002} Pfahl, E., Rappaport, S., Podsiadlowski, P., 2002, \apjl, 571, 37 \bibitem[Stark et al.(2004)]{stark2004} Stark, A. A., et al., 2004, \apjl, 614, 41 %%Stark, A. A., Martin, C. L., Walsh, et al., 2004, \apjl, 614, 41 \bibitem[Zhao et al.(2003)]{zhao2003} Zhao P., et al., 2003, BAAS., 7, 37.04 \end{thebibliography} %%\clearpage %% The following command ends your manuscript. LaTeX will ignore any text %% that appears after it. \end{document}