A LIMIT ON THE NUMBER OF ISOLATED NEUTRON STARS DETECTED IN THE ROSAT ALL-SKY-SURVEY BRIGHT SOURCE CATALOG

The X-ray source class of isolated neutron stars (INSs) is observationally defined by their high X-ray to optical flux ratio, F_X/F_opt ≳ 10⁴ (Treves et al. 2000). At discovery, these objects do not exhibit radio pulsations and are not associated with supernova remnants (SNRs).

The study of this phenomenological class ultimately promises a more complete picture of neutron star properties and evolution. Following formation in a supernova (SN) explosion, the "standard" neutron star (that is, one with a core composed of beta-equilibrium nuclear matter) evolves thermally, cooling to a temperature T ∼ 10⁶ K after t ∼ 10⁶ yr (Page et al. 2004). The cooling is modestly affected by the presence of strong (B ∼ 10¹⁵ G) magnetic fields (Arras et al. 2004). During this period, if the neutron star is emitting isotropically in the X-ray band, detection and study of the source will be a straightforward issue of improving the flux sensitivity of X-ray instrumentation. This simple picture could be altered if neutron stars, at birth, have strong toroidal magnetic fields, which may significantly modify the isotropy of their early emission (Page et al. 2007).

Over 1500 radio pulsars have been discovered (Manchester et al. 2005); and a few hundred low-mass-X-ray binaries (LMXBs) and high-mass-X-ray binaries (HMXBs) are cataloged (Liu et al. 2006, 2007). However, star formation models predict N_MW ∼ 10⁹ neutron stars have formed within the Milky Way (Timmes et al. 1996), implying that the vast majority of stellar evolution remnants have not yet been observed.

While one can attempt to infer the natal properties of neutron stars observed as radio pulsars and X-ray binaries, selection effects are present in both populations. Radio pulsars exhibit significant surface dipolar magnetic fields (B ∼ 10⁸–10¹³ G; Camilo et al. 1994, 2000); the distribution of magnetic field strength and geometry at birth, and their subsequent time evolution, are not strongly observationally constrained; it may be that only a fraction of neutron stars are born with such magnetic fields, implying that relatively few NSs are ever observable as radio pulsars.

The uncertain emission mechanism(s) of radio pulsars have made it challenging to quantify the observational selection effects which impact their detection in radio surveys. Nonetheless, recent analyses modeling this class conclude that the properties of known pulsars and pulsar surveys, including best estimates of the above selection effects, can account for all neutron stars observed (Faucher-Giguère & Kaspi 2006).

Observation and characterization of the INS population (which, in the present work, will include all neutron stars which are first identified via their high F_X/F_opt) provides a means to more fully characterize the neutron star population of the Galaxy, without selections related to magnetic field strength, geometry, evolution, and the various phases (and consequences) of binary stellar evolution.

Modeling of the INS population is, however, susceptible to theoretical uncertainties in neutron star cooling mechanisms, including the possibility of enhanced neutrino emissivity due to hyperon condensates (Page et al. 2004). Determining the ages (to ≲50%) of individual isolated thermally emitting neutron stars, independent of the observed surface temperature, is currently only possible for objects with precision parallax and proper-motion measurements (e.g., Kaplan et al. 2002, 2007), which allow their space velocity to be back-extrapolated to a likely birthplace in an OB-association.

In the absence of accurate ages for individual INSs, comparing individual INS properties with cooling models is effectively an interpretive act rather than a confrontation of theory with data. To test these models, therefore, will ultimately require a sufficient number of observed INSs such that models of their birth properties, evolution, and dynamics can be challenged to reproduce the observed population, as is now usefully done for the radio pulsar population (Cordes & Chernoff 1998; Arzoumanian et al. 2002; Gonthier et al. 2004; Faucher-Giguère & Kaspi 2006).

INSs evolve in the absence of influence of a companion (such as accretion of ∼0.1–0.3 M_☉ over the lifetime of an LMXB), and so offer an opportunity to study the physics of compact objects directly—their formation, dynamics, thermal evolution, and magnetic field evolution. Initial estimates (Blaes & Madau 1993) for the number of INSs which would be detected in the ROSAT All-Sky-Survey Bright Source Catalog (RASS/BSC; Voges et al. 1999) overestimated the number which were eventually observed, by a factor of ∼100 (as discussed previously; Rutledge et al. 2003, R03 hereafter). This overestimation was due to the fact that the observable population's luminosity was thought to be dominated by simple Bondi–Hoyle accretion, which scales approximately with the neutron star velocity ∝v⁻³. However, the velocity distribution for radio pulsars was later found to underestimate the typical velocity by a significant factor, which subsequently altered the conclusions by decreasing both the luminosity and the number of detectable INSs. Later estimates, which employ results of magnetohydrodynamic (MHD) simulations of Bondi–Hoyle accretion, found that the modified Bondi formula dramatically suppresses the observable population even further, to the extent that none are expected detected above RASS/BSC sensitivity (Perna et al. 2003). Thus, the interpretation of INSs observed from the RASS/BSC has been as post-natal cooling from neutron stars (Popov et al. 2000, 2003, 2006b, 2006a; Popov 2001; Posselt et al. 2007, 2008).

The present work continues our search for INSs—X-ray-bright sources with no observed off-band emission at discovery. This follows our previous work (R03), using essentially the same selection methods for INS candidates detected by the RASS/BSC. We have expanded the selection of INS candidates based on P_no-id—the probability that the RASS/BSC X-ray source is not associated with any off-band counterparts in the catalogs considered (USNO-A2; NVSS and IRAS)—from P_no-id ⩾ 0.90 to P_no-id ⩾ 0.8. Doing so increases the fraction of all INSs detected within the analyzed region of the RASS/BSC (δ ⩾ −39°, due to the sky coverage of NVSS) which pass our selection, from ∼20% (R03) to 30% (in the present work); this insures a greater number of INSs detected in the RASS/BSC end up in our selected INS candidates list.

To find 100% of the INSs within the RASS/BSC, an analysis such as the one performed here would be required for all 18,806 RASS/BSC X-ray sources. The present analysis makes use of a statistical selection which reduces the number of sources to be analyzed by a factor of ×100 (from 18,806 RASS/BSC sources to 147 sources), and will permit identification of almost 1/3 of all INSs among RASS/BSC sources. To identify 100% of the INSs in the RASS/BSC sources thus will require approximately 100× the amount of observations analyzed here, to obtain a factor of 3 more INSs. Thus, searches for INSs and classification of their populations using the techniques described here involve a balance between observational resources invested (in Swift and Chandra observations), and the number of INSs which will be identified. The size of the sample analyzed here is a compromise between these two goals, providing a significant improvement over previous characterization of the INS population in the RASS/BSC (R03).

We also report follow-up observations with Swift, as well as archival observations with Chandra, XMM-Newton, and ROSAT/High Resolution Imager (HRI), which permit localization and association with off-band (optical, IR and UV) sources.

As previously, we note that the present analysis produces an upper limit not only on objects that follow the strict definition of an INS–non-radio pulsar, non-magnetar, and non-SNR associated objects—but of high X-ray/optical flux ratio-selected populations of all types (such as anomalous X-ray pulsars (AXPs)). It should be noted that in our analysis, we may refer to objects as "INS" based on their X-ray/optical flux ratio and not on other properties that may technically exclude it from the phenomenological class. As an example, one object in our selection sample, 1RXS J141256.0+792204 (also known as Calvera) was identified as an isolated compact object of some kind based on its X-ray/optical flux ratio (Rutledge et al. 2008). Although more observations are necessary to determine its classification as either an INS, AXP, or radio pulsar, we classify it in our analysis as an "INS."

The analysis proceeds in three parts:

• 1.  
  Selection of the candidate INS sources from the RASS/BSC and off-band catalog cross-correlations, to find X-ray sources which are not likely to be associated with off-band counterparts.
• 2.  
  Follow-up observations (with Swift/XRT + UVOT); literature search for well-observed and understood counterparts to the RASS/BSC sources; and archival analysis of Chandra (06 systematic positional uncertainty,³ 1σ), XMM-Newton (2'' systematic positional uncertainty,⁴ 1σ), and ROSAT/HRI (63 systematic positional uncertainty⁵) which permit refined localization of the RASS/BSC X-ray source to search for off-band associations. We consider two point sources to be spatially associated if the probability that an as bright or brighter off-band cataloged object should lie as close or closer to the RASS/BSC X-ray source is ⩽1%; that is, the probability of chance association between the RASS/BSC source and the optical source (P_chance) is ⩽10⁻². For the number of RASS/BSC sources in our sample (147), this will produce no more than ∼1–2, on average, false associations with off-band counterparts. With regards to literature searches: some RASS/BSC sources have multiple X-ray observations in which phenomena (such as type-I X-ray bursts) are observed which are not observed from INSs, while others do not. We offer references for work which determines the object is "not an INS," or we otherwise conclude that the object is classified as "undetermined."
• 3.  
  Following the same statistical procedure as employed in our previous work (R03), we calculate a new upper limit on the number of INSs in the RASS/BSC.

In the following subsections, we provide further detail on each of these steps.

2.1. Selecting the RASS/BSC INS Candidate Sample

We draw candidate sources from the same selection as our previous analysis (R03), in which a full description of this selection is found. In brief, we performed a statistical cross-identification between the RASS/BSC and the USNO-A2 (optical), IRAS (infrared), and NVSS (radio) catalogs, quantifying (among other things) the probability P_no-id that each RASS/BSC X-ray source was not associated with any of the off-band catalog sources.

To calculate P_no-id  all sources from USNO-A2 within 75'' and from IRAS and NVSS within 150'' of the RASS/BSC positions were used to determine a likelihood of association for every X-ray source i and off-band source j from catalog C. This likelihood ratio is given by

$$\begin{equation} {\rm LR}_{i,j;C} = \frac{\mbox{exp}\big(-r^2_{i,j}/2\sigma ^2_{i,j}\big)}{\sigma _{i,j}N(>F_j;C)}, \end{equation} \tag{ 1 }$$

where r_i,j is the separation between the X-ray and off-band source, σ_i,j is the uncertainty in r, and N(>F_j; C) is the number of sources in the catalog C with fluxes F greater than off-band source j.

Background fields, which are 24 off-source positions set in a 5×5 grid around each X-ray source position, consist of circles with radii of 75'' separated by 150'' for USNO-A2, and circles with radii of 150'' separated by 300'' for IRAS and NVSS. The probability that a candidate counterpart j is spatially associated with the X-ray source and is not a background object is given by

$$\begin{equation} R_{i,j}({\rm LR}_{i,j;C}) = \frac{N_{{\rm src}}({\rm LR}_{i,j;C})-N_{{\rm bkg}}({\rm {\rm LR}}_{i,j;C})}{N_{{\rm src}}({\rm LR}_{i,j;C})}, \end{equation} \tag{ 2 }$$

where N_src(LR_i,j;C) is the number of objects in the source fields that have a value of LR within some δLR of LR_i,j;C and N_bkg(LR_i,j;C) is the same but for the background fields.

Thus, we can obtain the probability that none of the objects in the field are associated with the X-ray source, given by

$$\begin{equation} P_{i,{\rm no-id}} = \frac{\Pi _j(1-R_{i,j})}{K}, \end{equation} \tag{ 3 }$$

where K is a normalization factor. A high value of P_no-id for a RASS/BSC X-ray source means it is unlikely that the RASS/BSC source is associated with any of the off-band catalog sources.

For the purpose of the analysis, 150 "control" X-ray sources were inserted among the source fields, distributed randomly in proportion to the local X-ray source density. Since they should have no detectable off-band counterparts, they behave like INSs. Thus, we are able to evaluate the statistical efficiency of our selection, based on the number of control sources that remain above our P_no-id cutoff.

We expand upon the earlier selection in the following two ways:

• 1.  
  We now include RASS/BSC sources with P_no-id > 0.8 (versus P_no-id > 0.9 in our previous work); in addition to increasing the number of candidate sources to be investigated, this increases the number of control sources which end up in our selection. Previously, 29 of 150 (19%) control sources ended up in our selection; here, 46 of 150 (31%). Because control sources have the same behavior expected from INSs (i.e., have no counterparts in the off-band catalogs), this increases the number of INSs on our candidate list from 19% of all INSs detected by the RASS/BSC in our survey region to 31%. Thus, we expect almost 1/3 of all INSs detected in the RASS/BSC in our survey region are in this selection.
• 2.  
  In our previous work, we examined only RASS/BSC sources with a hardness ratio HR1 < 0, indicating an X-ray spectrum consistent with an effective temperature kT_eff < 200 eV.⁶ In the present work, we include all RASS/BSC sources regardless of their spectrum. This is due to the fact that, in the interim, we have discovered at least one INS with a spectrum harder than this limit (Rutledge et al. 2008), as well as the existence of the magnetars, a class of INS with kT_eff ∼ 600 eV. We therefore impose no spectral constraints on our RASS/BSC INS candidates.

We also no longer impose the requirement that the X-ray flux history be consistent with no variability: there are now a handful of transient magnetars (Ibrahim et al. 2004; Israel et al. 2007; Kumar & Safi-Harb 2008), whose average flux histories, absent giant flares, and microflares, still vary by 2–3 orders of magnitude—brightening on timescales of ∼ hours to days, and subsequently fading over timescales of days to weeks (in some cases, months).

The RASS/BSC spatial extendedness parameter, which characterizes X-ray source extendedness with a likelihood parameter extl = −log(P) where P is the probability of the analysis concluding source extendedness from a point source, is not used as a selector in this work.

A brief analysis has raised the possibility of false positives for source extendedness in the RASS/BSC catalog. There are, at present, 105 RASS/BSC sources identified as isolated white dwarfs (WDs) according to the SIMBAD⁷ database. Of these, 21 are listed with extl>10, corresponding to a chance detection from a non-extended source of 4×10⁻⁵; we would expect only 0.005 such detections, on average, in a population size of 105.

Of the 21 isolated WDs with extl>10, one has been observed with Chandra on-axis (<1' offset) and without a grating: 1RXS J103210.2+532941. We find no evidence that that WD is extended at Chandra resolution in the course of this 3 ks observation. Thus, we conclude that the extl parameter does not accurately represent the likelihood of spatial extent in the 0.1–2.4 keV photon band for sources in the BSC. (The alternative scenario, that these WDs are associated with extended X-ray emission that is variable on ∼year timescales, has not otherwise been considered.)

Furthermore, there are two known observational classes of neutron stars which are associated with SNRs, the compact-central objects (CCOs), and magnetars; since SNRs can be spatially extended at ROSAT/PSPC spatial resolution, discarding candidate INSs on the basis of spatial extent would miss these classes of sources. Given the systematic uncertainty in the accuracy of the extl parameter, we do not use spatial extent as a cut for candidate INSs in the RASS/BSC.

Thus, we make no secondary cuts of INS candidates from the RASS/BSC source selection other than the single selection criterion based on P_no-id. We find 147 RASS/BSC sources with P_no-id > 0.80, which are listed in Table 1. So that we may compare results with previous work (R03), we note that of these 147 RASS/BSC sources, 71 have hardness ratio HR1 < 0, implying a spectrum consistent with a thermal spectrum of kT_eff < 200 eV.

Table 1. Candidates and Identifications

1RXS J	Pno-id	Type	Name	Observed?a	Hardness Ratio	ID
J001832.0+162634	0.872	Cluster of galaxies	ClG 0015.9+1609	CXH	0.73	Not an INS
J003854.9−034252	0.818	...	...	S	0.56	Not an INS
J004330.9+411452	0.865	...	2E 0040.8+4057	CXHS	0.63	Not an INS
J004704.8−204743	0.855	...	...	XH	1.00	Undetermined
J012428.1−335504	0.861	High proper-motion star	G 269-153	S	−0.14	Not an INS
J013653.5−351012	0.884	Seyfert 1 galaxy	2MASSI J0136544-350952	XHS	−0.81	Not an INS
J014205.0+213045	0.858	Cluster of galaxies	ClG J0142+2131	S	1.00	Not an INS
J015311.6−210545	0.876	High proper-motion star	G 274-113	S	−0.14	Not an INS
J020146.5+011717	1.000	Star	RX J0201.7+0117	S	−0.25	Not an INS
J020210.5−020616	0.911	...	...	S	0.04	Undetermined
J020317.5−243832	0.939	...	...	C	−0.16	Not an INS
J020503.6−173717	0.815	...	...	S	−0.52	Undetermined
J022152.6+281047	0.833	...	RX J0221.8+2810	S	−0.15	Not an INS
J024528.9+262039	0.929	Star	...	C	−0.19	Not an INS
J024814.8−193954	0.808	...	...	S	−0.19	Not an INS
J024946.0−382540	1.000	...	...	S	−0.24	Not an INS
J025414.5+413530	0.801	...	...	CXH	0.99	Not an INS
J031413.7−223533	0.925	Nova-like star	V* EF Eri	XS	−0.61	Not an INS
J032620.8+113106	1.000	T Tau-type star	[LHC2000] J032621.24+113055.9	S	−0.16	Not an INS
J033642.5−095506	0.816	...	...	S	0.71	Undetermined
J034414.1+240623	0.939	Star in cluster	BD+23 502	H	0.79	Not an INS
J035101.5+141404	0.833	...	...	S	−0.35	Not an INS
J040314.6−360927	0.832	Star	2E 0401.4−3617	S	−0.22	Not an INS
J040543.6−282114	0.807	...	...	S	−0.24	Undetermined
J040814.3+400723	0.811	...	...	S	−0.14	Not an INS
J040817.9+294951	0.825	...	...	S	1.00	Not an INS
J040913.8+110833	0.905	...	...	S	1.00	Not an INS
J041215.8+644407	1.000	Flare star	GJ 3266	S	−0.51	Not an INS
J042513.4+171548	0.850	White dwarf	V* V805 Tau	S	−0.30	Not an INS
J043334.8+204437	0.925	Flare star	GJ 3296	S	−0.15	Not an INS
J044048.0+292440	0.914	...	...	S	0.93	Undetermined
J050909.9+152740	0.890	Flare star	GJ 3335	S	−0.13	Not an INS
J051028.9+022051	0.805	...	...	S	0.67	Not an INS
J051315.9+025227	0.925	...	...	S	0.37	Not an INS
J051319.1+013525	0.855	...	...	S	−0.14	Not an INS
J051354.0+023722	0.816	...	...	S	0.00	Not an INS
J051541.7+010528	0.961	Cataclysmic Var. AM Her type	V* V1309 Ori	XH	−0.92	Not an INS
J051723.3−352152	0.912	High proper-motion star	L 449-1	S	−0.18	Not an INS
J051854.5+323827	1.000	...	...	S	0.06	Not an INS
J052929.6+031838	0.805	...	...	S	0.69	Not an INS
J053510.8−044850	1.000	Nebula of unknown nature	...	CXH	1.00	Undetermined
J053516.3−044033	0.903	...	2E 0532.8−0442	CXH	1.00	Not an INS
J053842.4−023525	0.815	...	...	CXH	0.51	Not an INS
J054042.8−020533	0.808	...	...	CXH	0.67	Undetermined
J054045.7−021119	1.000	Variable star	V* V611 Ori	XH	0.82	Not an INS
J055228.1+155313	0.815	White dwarf	GD 71	...	−0.99	Not an INS
J055734.5−273534	0.889	...	...	S	0.46	Undetermined
J055800.7+535358	0.849	Cataclysmic Var. DQ Her type	V* V405 Aur	XH	−0.53	Not an INS
J060452.1−343331	0.961	Eruptive variable star	V* AP Col	H	0.06	Not an INS
J062203.6+744106	0.826	...	...	S	0.32	Not an INS
J063354.1+174612	0.871	Pulsar	SN 437	CXH	−0.92	INS
J074451.8+392733	0.813	...	...	C	0.83	Not an INS
J075523.9+372634	0.907	Possible BL Lac	SDSS J075523.11+372618.8	S	0.96	Not an INS
J075556.7+832310	0.904	Flare star	GJ 1101	S	−0.21	Not an INS
J082124.5−362848	0.847	...	...	S	0.52	Not an INS
J082355.1+394745	0.832	...	RX J0823.9+3947	S	0.77	Not an INS
J084127.7−102843	0.836	...	...	S	1.00	Undetermined
J085247.0+223040	0.961	...	RX J0852.7+2230	X	0.13	Not an INS
J090137.2−052341	0.815	...	...	S	−0.11	Not an INS
J090717.4+225254	0.849	Star	HD 78141	S	−0.06	Not an INS
J091010.2+481317	0.931	Seyfert 1 galaxy	QSO B0906+484	S	−0.43	Not an INS
J091112.2+174634	0.849	...	RX J0911.2+1746	C	0.46	Not an INS
J094432.8+573544	1.000	BL Lac-type object	SDSS J094432.32+573536.0	HS	−0.39	Not an INS
J094454.2−122047	0.804	High proper-motion star	G 161-71	S	−0.15	Not an INS
J094831.1−333814	0.801	...	...	S	0.73	Undetermined
J101628.3−052026	0.835	White dwarf	GSC 04910-01132	...	−1.00	Not an INS
J102213.7+735433	0.803	...	RX J1022.2+7354	S	0.09	Not an INS
J102954.3+614732	0.896	...	RX J1029.9+6147	H	−0.29	Not an INS
J103347.4−114146	0.814	...	...	...	−0.96	Undetermined
J104710.3+633522	0.939	Cataclysmic Var. DQ Her type	V* FH UMa	X	−0.97	Not an INS
J110521.6−073525	0.870	...	RX J1105.3−0735	S	−0.13	Not an INS
J112430.5+435131	0.832	...	RX J1124.5+4351	S	0.84	Not an INS
J115309.7+545636	1.000	...	RX J1153.1+5456	C	−1.00	Undetermined
J120711.0+364745	0.876	Seyfert 1 galaxy	2MASS 12071782+3648008	S	−0.03	Not an INS
J121732.6+152843	0.961	...	RX J1217.5+1528	S	0.45	Not an INS
J121900.7+110727	0.808	...	RX J1219.0+1107	S	−0.25	Not an INS
J122308.4+110054	0.953	...	RX J1223.1+1100	S	0.48	Not an INS
J122940.6+181645	1.000	BL Lac-type object	[ZEH2003] RX J1229.6+1816 1	CS	−0.38	Undetermined
J123319.0+090110	0.933	Star in double system	GJ 473 B	...	−0.43	Not an INS
J124849.0+333454	0.845	...	RX J1248.8+3334	S	−0.48	Not an INS
J125015.2+192357	1.000	Seyfert 1 galaxy	[ZEH2003] RX J1250.2+1923 1	S	−0.64	Not an INS
J125721.8+365431	0.810	...	RX J1257.3+3654	S	0.79	Undetermined
J125947.9+275636	0.961	Cluster of galaxies	ACO 1656	CXHS	0.33	Not an INS
J130034.2+054111	0.954	Flare star	V* FN Vir	S	−0.16	Not an INS
J130205.2+155122	1.000	...	RX J1302.0+1551	S	0.13	Undetermined
J130402.8+353316	1.000	Quasar	QSO B1301+358A	S	−0.34	Not an INS
J130547.2+641252	1.000	Star	RX J1305.7+6412	XH	−0.82	Undetermined
J130631.3+192229	1.000	...	RX J1306.5+1922	S	0.11	Not an INS
J130753.6+535137	0.939	Cataclysmic Var. AM Her type	V* EV UMa	XH	−0.90	Not an INS
J130848.6+212708	1.000	Neutron star candidate	RX J1308.8+2127	CXH	−0.20	INS
J131011.9+474521	0.960	High proper-motion star	LHS 2686	S	0.25	Not an INS
J132041.2−030010	0.815	...	...	S	−1.00	Undetermined
J132833.1−365425	0.961	High proper-motion star	...	C	−0.57	Not an INS
J133032.3+720931	0.961	...	RX J1330.5+7209	S	0.18	Not an INS
J133825.0−251634	0.872	...	[FS2003] 0678	S	0.10	Not an INS
J134210.2+282250	1.000	Cataclysmic variable star	RX J1342.1+2822	CH	−0.95	Not an INS
J134619.1−141834	0.960	...	...	S	0.09	Not an INS
J135152.7+462149	0.925	Cluster of galaxies	RXC J1351.7+4622	S	0.35	Undetermined
J140818.1+792113	0.890	...	RX J1408.3+7921	S	−0.88	Undetermined
J141256.0+792204	0.904	...	RX J1412.9+7922	CS	0.50	INS
J141703.1+314249	0.812	...	RX J1417.0+3142	S	−0.44	Not an INS
J142644.1+500633	0.914	...	RX J1426.7+5006	S	−0.85	Undetermined
J143652.6+582104	0.885	...	RX J1436.8+5821	S	−0.31	Not an INS
J144359.5+443124	0.903	...	RX J1443.9+4431	S	0.37	Undetermined
J145010.6+655944	0.947	Cataclysmic variable star	RX J1450.1+6559	C	−0.83	Not an INS
J145234.9+323536	1.000	...	RX J1452.5+3235	C	0.00	Undetermined
J145729.4+083356	0.804	Seyfert 1 galaxy	RX J1457.4+0833	S	−0.21	Not an INS
J153840.1+592118	0.961	LINER-type active galaxy nucleus	NGC 5982	S	0.73	Not an INS
J160518.8+324907	1.000	Star	RX J1605.3+3249	CXH	−0.70	INS
J161455.4−252800	0.873	...	...	CH	0.95	Not an INS
J162721.6−244144	0.925	Variable star of orion type	V* V2247 Oph	CXH	0.69	Not an INS
J163212.8−244013	0.820	Variable star of orion type	V* V2248 Oph	H	1.00	Not an INS
J163421.2+570933	0.961	Variable of BY Dra type	V* CM Dra	XH	−0.30	Not an INS
J163910.7+565637	1.000	Active galaxy nucleus	RX J1639.1+5656	C	−0.13	Not an INS
J164020.0+673612	0.892	Flare star	GJ 3971	S	−0.12	Not an INS
J165344.7+101159	0.905	...	RX J1653.7+1011	S	0.72	Undetermined
J171502.4−333344	0.961	...	...	S	0.19	Undetermined
J172148.4−051729	0.839	...	...	S	0.92	Undetermined
J173006.4+033813	0.860	...	...	S	0.62	Not an INS
J173157.7−335007	0.830	Low-mass-X-ray binary	Slow burster	CXS	1.00	Not an INS
J173253.6−371200	1.000	...	...	...	0.02	Undetermined
J173319.3−255416	0.961	...	...	XH	1.00	Not an INS
J173546.9−302859	0.864	Low-mass-X-ray binary	...	CH	1.00	Not an INS
J180132.3−203132	0.944	Low-mass-X-ray binary	X Sgr X-3	CXH	0.99	Not an INS
J181506.1−120545	0.846	Low-mass-X-ray binary	4U 1812−12	CHS	1.00	Not an INS
J182102.0−161309	0.915	...	...	CX	0.74	Undetermined
J183543.6−325928	0.823	Low-mass-X-ray binary	RX J1832-33	CXH	0.93	Not an INS
J185557.3+233410	0.820	Variable of BY Dra type	V* V775 Her	...	−0.11	Undetermined
J191426.1+245641	0.843	Cataclysmic variable star	V* V407 Vul	CXHS	0.95	Not an INS
J204249.0+412242	0.805	Flare star	V* V1589 Cyg	S	−0.22	Undetermined
J205549.4+435216	0.857	...	...	S	0.92	Not an INS
J210324.7+193026	0.838	...	...	S	0.62	Not an INS
J212700.3+101108	0.832	...	...	S	0.50	Undetermined
J213944.3+595016	0.821	...	...	XS	0.90	Not an INS
J214918.6+095130	0.884	...	RX J2149.3+0951	S	0.43	Not an INS
J221144.6−034947	0.804	...	2E 2209.1−0404	C	0.86	Undetermined
J221403.0+124207	0.821	Dwarf nova	V* RU Peg	X	0.48	Not an INS
J223832.0−151809	0.810	Flare star	V* EZ Aqr	...	−0.50	Not an INS
J230334.0+152019	0.925	...	RX J2303.5+1520	H	1.00	Undetermined
J230340.4−352420	0.845	...	...	S	−0.95	Undetermined
J231543.7−122159	0.950	High proper-motion star	LHS 3918	S	−0.13	Not an INS
J231728.9+193651	0.904	Double or multiple star	[ZEH2003] RX J2317.4+1936 3	S	−0.40	Not an INS
J233407.2−033533	0.860	...	...	...	0.80	Undetermined
J233757.2+271031	0.831	...	RX J2337.9+2710	XH	1.00	Undetermined
J234305.1+363226	0.868	...	...	S	−0.44	Not an INS
J234421.2+213601	0.876	...	RX J2344.3+2136	S	−0.10	Not an INS
J234836.5−273935	0.829	...	...	S	−0.32	Not an INS

Note. ^aFacilities: C, Chandra; X, XMM; H, ROSAT; S, Swift

Download table as:  ASCIITypeset images: 1 2 3

2.2. Follow-up Swift Observations and X-ray Archival Data Analysis

2.3. Individual Objects

Here, we use the same calculation as previously (see Section 5 of R03) to derive upper limits on the number of INSs detected in the full-sky RASS/BSC. We refer the reader to that reference for the description of this calculation. In short, with knowledge of the selection efficiency of INSs detected in the RASS/BSC to our sample; the number of sources in our sample which are known to be INSs; the number of sources in our sample which are known to be not an INS; and the number of sources in our sample which are of undetermined type (INS, or not), one can place an upper limit on the total number of INSs in the RASS/BSC.

To estimate the upper limit on the number of INSs in our catalog, we use the following values, where all variables take the same meaning as in R03. Of A = 150 control sources, B = 46 passed our P_no-id selection; this implies that a real INS has probability p of passing our INS selection, which can be represented as a binomial distribution:

$$\begin{equation} P_{\rm XID}(p) = \frac{ p^{B}(1-p)^{A-B}}{\int _0^1 p^{B}(1-p)^{A-B}\; dp}. \end{equation} \tag{ 4 }$$

From the 15,205 RASS/BSC sources above our declination cut, a total T = 147 passed the P_no-id selection. Based on the fraction of control sources which passed our P_no-id selection (46/150 = 30.67%), we expect that 30.67%, on average, of all INSs among the 15,205 RASS/BSC sources are in our selected sample of T = 147 RASS/BSC sources.⁸

After examining the literature, Swift observations, and archival data analysis, there are N_INS,min = 4 objects which we count as INSs (1RXS J063354.1+174612 which is also known as Geminga; 1RXS J130848.6+212708; 1RXS J141256.0+792204 which is also known as Calvera; and 1RXS J160518.8+324907), 107 sources which are not an INS, and BG =36 sources are found to be of "undetermined" type. With these values, we can determine the unnormalized probability P_BG(N) that N sources in our selection of 147 are INSs:

$$\begin{equation} P_{\rm BG}(N) = { {\frac{ \big(\begin{array}{@{}c@{}} \ssty N \\ [-3pt] \ssty N_{\rm INS, min}\end{array}\big) \big(\begin{array}{@{}c@{}}\ssty T - N \\[-3pt] \ssty T-{\rm BG}-N_{\rm INS, min}\end{array} \big) }{ \big(\begin{array}{@{}c@{}} \ssty T \\ [-3pt] \ssty T-{\rm BG}\end{array} \big) }},} \end{equation} \tag{ 5 }$$

where N_INS,min ⩽ N ⩽ N_INS,max and N_INS,max is the maximum number of INSs that could be present in our sample (N_INS,min + BG = 40).

Combining Equations (4) and (5) yields the unnormalized probability P_INS(M') that the total number of INSs in our survey field is M':

$$\begin{eqnarray} &&P_{\rm INS, unnormalized}(M^{\prime }) = \sum _{N = N_{\rm INS, min}}^{{\rm min}(M^{\prime }, N_{\rm INS, max})} \Bigg(\frac{P_{\rm BG}(N)}{\sum _{N = N_{\rm INS, min}}^{{\rm min}(M^{\prime }, N_{\rm INS, max})} \: P_{\rm BG}(N) } \nonumber\\ &&\qquad\times \int _0^1 P_{\rm XID}(p)\, \frac{p^N (1-p)^{M^{\prime }-N} }{\int _0^1 p^N (1-p)^{M^{\prime }-N}\; dp}\; dp\Bigg). \end{eqnarray} \tag{ 6 }$$

Finally, to obtain the probability that there are ⩾M INSs in our sample, we sum

$$\begin{equation} P_{\rm INS}(\ge M) = \sum ^\infty _{M^{\prime } = M} \frac{\: P_{\rm INS, unnormalized}(M^{\prime })}{\left(\sum ^\infty _{M^{\prime } = N_{\rm INS, min}} \:P_{\rm INS, unnormalized}(M^{\prime }) \right)}. \end{equation} \tag{ 7 }$$

Using the above values, we place a 90% confidence upper limit on the number of INSs among the 15,205 in our survey region to be ⩽39 and a 99% confidence upper limit of ⩽57. Rescaling this to the total number of RASS/BSC sources full sky, these correspond to full-sky upper limits on the number of INSs in the RASS/BSC of ⩽48 (90% confidence) and ⩽70 (99% confidence).

Variability. Of the 36 RASS/BSC sources in our selection which we classify as "undetermined," 25 were undetected in a second-epoch X-ray image (17% of the total sample), which implies a significant decrease in the X-ray flux. While some classes of neutron stars (e.g., magnetars) can exhibit significant fading, we can calculate the number of INSs from classes which are not variable, by classifying these 25 faded sources as "not an INS," and performing the calculation again. Doing so, we find the full-sky upper limit on the number of non-variable INSs in the RASS/BSC to be ⩽39 (90% confidence) and ⩽56 (99% confidence).

Spectrally soft, non-variable sources. In previous work (R03), only soft (kT_eff < 200 eV) sources which exhibit no significant variability were considered as INSs. In doing so, full-sky upper limits on the number of soft, non-variable INSs to be ⩽67 (90% confidence) and ⩽104 (99% confidence) were derived. In the present analysis, we have 71 spectrally soft sources; of these, 3 are identified as an INS, 66 are not an INS, and 2 are undetermined. With these, we place new full-sky upper limits on the number of soft, non-variable INSs in the RASS/BSC of ⩽31 (90% confidence) and ⩽46 (99%) confidence, both a little more than a factor of 2 lower than the previously derived limits.

The above results are summarized in Table 4.

3.1. A Targeted INS Candidate List

We find X-ray objects which were detected in the RASS/BSC, with localizations ∼13'' (1σ), which were redetected with Swift/XRT follow-up observations and localized with ∼35, and which still have no likely counterparts in USNO-A2, Two Micron All Sky Survey (2MASS), NVSS, IRAS, or in contemporaneous UVOT observations.

We are in the process of targeting these sources in Chandra observations, to bring the positional uncertainties down ⩽06, and deep optical observations, to demonstrate F_X/F_opt flux ratio >1000, which has been a robust indicator of an INS. We list these sources in Table 5.

Table 5. Isolated Neutron Star Candidates Detected in Second-Epoch Observations

1RXS J	Pno-id	Flux (×10−13  erg $\mbox{$\rm \,cm$}^{-2}$ s−1)
J044048.0+292440	0.914	36.6
J084127.7−102843	0.836	11
J130205.2+155122	1.000	3
J144359.5+443124	0.903	8.5
J171502.4−333344	0.961	32.5
J212700.3+101108	0.832	15.1
J213944.3+595016	0.821	45
J230334.0+152019	0.925	6.9
J233757.2+271031	0.831	4.5

Download table as:  ASCIITypeset image

3.2. The Relevant Populations, and eROSITA

We have placed upper limits on the number of INSs—independent of spectral and variability properties—in the RASS/BSC, of ⩽48 (90% confidence) and ⩽70 (99% confidence). When we limit ourselves to spectrally soft, non-variable INSs, these limits are ⩽31 (90% confidence) and ⩽46 (99% confidence)—a factor of 2 lower than limits derived previously (R03).

Using the same analysis as in previous work (Section 5.2, R03), we can place a limit on SN rate of progenitors to these objects, using a naive and optimistic model for the resulting sources. The following assumptions are made for this model:

• 1.  
  An INS production rate of γ₋₂ per 100 yr.
• 2.  
  INSs are produced in a flat disk with constant SNe rate per unit area, out to a radius of R_disk = 15 kpc.
• 3.  
  INSs are produced velocities of <10 kpc Myr^-1 perpendicular to this disk.
• 4.  
  The resulting INSs maintain a luminosity 2 × 10³²  erg s⁻¹ for a period of τ = 10⁶ yr in the ROSAT/PSPC passband of 0.1–2.4 keV.
• 5.  
  The INSs exhibit an α = 3 spectrum.
• 6.  
  The limit on all INSs (assuming ⩽50) corresponds to a limit.
• 7.  
  The effects of absorption (which are likely to be important) are neglected (permitting INSs to be detected to a distance of 10.3 kpc).

We can then place a limit of γ₋₂ < 0.019 (90% confidence), corresponding to an INS birthrate of one per 5300 yr. If we assume a spatial average absorption of N_H = 3  × 10²¹ cm^-2 (limiting detection to sources within a distance of 1.5 kpc), the limit is γ_-2<0.8.

In the process of obtaining these limits, we have identified one new confirmed INS (Rutledge et al. 2008, Calvera). After second-epoch X-ray observations with Swift, we find nine INS candidates. We are presently following-up observationally on these candidates, with better X-ray localizations and deep optical imaging, to place limits of F_X/F_opt>1000, on these sources.

The authors would like to express their appreciation to Neil Gehrels, Jamie Kennea, and the Swift team for carrying out the fill-in target observations which form the basis for this paper. This publication makes use of data products from the Two Micron All Sky Survey, which is a joint project of the University of Massachusetts and the Infrared Processing and Analysis Center/California Institute of Technology, funded by the National Aeronautics and Space Administration and the National Science Foundation. This research has made use of the NASA/IPAC Infrared Science Archive, which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. The Digitized Sky Surveys were produced at the Space Telescope Science Institute under US Government grant NAGW-2166. The images of these surveys are based on photographic data obtained using the Oschin Schmidt Telescope on Palomar Mountain and the UK Schmidt Telescope. The plates were processed into the present compressed digital form with the permission of these institutions. The DPOSS project was generously supported by the Norris Foundation. R.E.R. and M.T. are supported by the Discovery Grants Program of the Natural Sciences and Engineering Research Council of Canada.

1RXS J012428.1−335504. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 13 ± 04); this is consistent with the classification for this in SIMBAD as a high proper-motion star.

1RXS J014205.0+213045. Examination of the Digitized Sky Survey (DSS) R-band images centered on the XRT position finds a group of ∼3 point sources with a surrounding nebulosity. The XRT source has been found to be extended, with probability that the observed X-ray brightness is consistent with the PSF of Swift/XRT of 10^−20.3; not an INS.

1RXS J015311.6−210545. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 21 ± 03); this is consistent with the classification for this in SIMBAD as a high proper-motion star.

1RXS J020146.5+011717. Examination of the DSS and 2MASS images of the counterpart reveals a possible optical/IR binary.

1RXS J024946.0−382540. The XRT source may also be associated with 2MASS 024945.8−382440.1 which may, itself, be a binary counterpart to 2MASS J024945.9−382536.6.

1RXS J031413.7−223533. This source has previously been classified as a nova/star (Watson et al. 1987; Beuermann et al. 1991), and classified as not an INS (R03).

1RXS J040314.6−360927. We note that the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 20 ± 03).

1RXS J040913.8+110833. The XRT source has been found to be extended, with probability that the observed X-ray brightness is consistent with the PSF of Swift/XRT of 10^−43.6.

1RXS J041215.8+644407. We note that the 2MASS and UVOT source, however, are significantly offset from one another (57±02).

1RXS J043334.8+204437. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 39 ± 03).

1RXS J044048.0+292440. No source is detected in the UVOT image (M2 > 21.5, 3σ). The XRT source has been found to be only marginally consistent with a point source, with probability that the observed X-ray brightness is consistent with the PSF of Swift/XRT of 10^−2.6; undetermined.

1RXS J050909.9+152740. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 35 ± 02).

1RXS J051315.9+025227. Swift/XRT observation on 2006 March 26 18:57 UT (observation ID = 00035532001) detects an apparently saturated X-ray source in the field, for which the standard astrometry is unreliable. Nonetheless, the best fit of this astrometry produces a position which is approximately 1' from the RASS/BSC position (the next nearest other RASS/BSC source is more than 15' away), but with an unquantifiable systematic uncertainty. The count rate appeared to be sufficiently great (>1 count s⁻¹) that the on-board rejection removed most of the X-ray counts from the datastream, which is in excess of the count rate which would be expected from the RASS/BSC count rate (0.09 counts s⁻¹). An XRT observation on 2006 March 29 04:52 UT (observation ID = 00035532002) redetects the X-ray source, at 05^h13^m159, + 02°51'43''. There is an optical source USNO 051317.4+025140.5 (B = 5.97, R = 3.81), 23'' away from the XRT source. SV* ZI 359 is located at 05^h13^m174, +02°51'40'' (Perryman et al. 1997). SV* ZI 359 is classified on SIMBAD as a K0.5III star, and a spectroscopic binary. At the X-ray flux from the RASS/BSC, the implied X-ray luminosity at the distance of SV* ZI 359 (105 ± 8 pc; Perryman et al. 1997) is L_X = 1.2×10³⁰  erg s⁻¹, which is consistent with the luminosity expected from the surface of such a star.

1RXS J051354.0+023722. We note that the XRT source is offset from the RASS/BSC position by 50''  and another XRT source with a lower count rate is located 160'' from the RASS/BSC position.

1RXS J051723.3−352152. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 19 ± 03); this is consistent with the classification for this in SIMBAD as a high proper-motion star.

1RXS J055734.5−273534. This source has been observed with Swift/XRT three times: on 2006 April 14 02:04:28 UT (observation ID = 00035513001, duration 769 s), on 2006 April 20 23:14:06 UT (observation ID = 00035513002, duration 1.1  ks), and on 2006 April 27 UT 06:22:55 (observation ID = 00035513003, duration 1.3  ks). For all three observations, no X-ray source was detected in the XRT field within 90'' of the RASS/BSC position (Shevchuk et al. 2009). Since <5 counts were detected during each observation, the upper limit on the flux is 2.1×10⁻¹³  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹ (0.1–2.4 keV).

1RXS J075556.7+832310. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 41 ± 02).

1RXS J094454.2−122047. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 26 ± 02); this is consistent with the classification for this object in SIMBAD as a high proper-motion star.

1RXS J120711.0+364745. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 10 ± 03).

1RXS J121900.7+110727. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 75 ± 03).

1RXS J122940.6+181645. This object was previously observed and undetected (R03).

1RXS J125947.9+275636. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 25 ± 0''). The XRT X-ray source appears extended to visual inspection, with an approximate width of 5' in diameter. This is consistent with the position for the associated RASS/BSC source. We conclude the XRT object is the RASS/BSC source, which corresponds to the Coma Cluster of galaxies.

1RXS J131011.9+474521. We note the 2MASS and UVOT sources are offset by 76±04, indicating a high proper-motion star, which is also apparent by comparing DSS B-band images from 1955 March 18 and 1995 March 27.

1RXS J132041.2−030010. This source has been observed with Swift/XRT twice: on 2006 June 10 00:05:06 UT (observation ID = 00035555001, duration 287 s) and on 2007 January 03 04:00:30 UT (observation ID = 00035555003, duration 1.3  ks). For both observations, no X-ray source was detected in the XRT field within 90'' of the RASS/BSC position.

1RXS J133825.0−251634. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 21 ± 03).

1RXS J141703.1+314249. Comparing archival DSS images (1950 April 12) with more recent 2MASS images show the identified 2MASS counterpart has exhibited significant proper motion over this epoch.

1RXS J142644.1+500633. To place the flux upper limit, we note detection of a marginal (3σ) source in the field, with a count rate of 25.7 counts ks⁻¹, which corresponds to a flux of 7.4×10⁻¹³  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.5–10 keV), which we adopt as a conservative flux upper limit.

1RXS J143652.6+582104. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 64 ± 04). A comparison of archival DSS images (1955 March 16) indicates the 2MASS source (observed 2000 March 3) exhibits significant proper motion over this epoch.

1RXS J153840.1+592118. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 14 ± 04).

1RXS J164020.0+673612. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 33 ± 03).

1RXS J171502.4−333344. Due to the large number of 2MASS point sources within the RASS/BSC error circle (11 2MASS sources within 16'' radius or 2σ), the probability that the RASS/BSC X-ray source is not associated with any 2MASS source is 10^−2.2. The most likely association was found to be with 2MASS J171502.19−333339.8 (J = 7.9), with probability of association P_id = 0.782 (Haakonsen & Rutledge 2009). The XRT source has been found to be only marginally consistent with a point source, with probability that the observed X-ray brightness is consistent with the PSF of Swift/XRT of 10^−2.0 (Shevchuk et al. 2009).

1RXS J205549.4+435216. We note that the 2MASS colors for this bright IR source would be unusual for a low-mass star (J = 6.36, H = 5.51, K = 5.04), and the object warrants IR spectral analysis; not an INS (Shevchuk et al. 2009).

1RXS J212700.3+101108. The XRT source has been found to be consistent with a point source, with probability that the observed X-ray brightness is consistent with the PSF of Swift/XRT of 0.40.

1RXS J213944.3+595016. The XRT source has been found to be consistent with a point source, with probability that the observed X-ray brightness is consistent with the PSF of Swift/XRT of 0.32

1RXS J230340.4−352420. This source has been observed with Swift/XRT three times: on 2006 January 05 12:04:41 UT (observation ID = 00035579001, duration 240 s), on 2006 April 19 00:32:11 UT (observation ID = 00035579002, duration 1.3  ks), and on 2006 May 1 03:13:03 TT (observation ID = 00035579003, duration 3.2  ks). For all three observations, no X-ray source was detected in the XRT field within 90'' of the RASS/BSC position.

1RXS J231543.7−122159. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 43 ± 04); this is consistent with the classification for this in SIMBAD as a high proper-motion star.

1RXS J231728.9+193651. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 34 ± 02).

1RXS J234421.2+213601. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 40 ± 03).

1RXS J234836.5−273935. We note the position of the 2MASS source is inconsistent with the position of the UVOT source (offset by 49 ± 04).

1RXS J001832.0+162634. This object has been identified as the galaxy cluster ClG 0015.9+1609, which has been observed with Chandra and found to be significantly extended (Ebeling et al. 2007); not an INS.

1RXS J004330.9+411452. Chandra/ACIS-I observation of M31 (observation ID = 1585, start time 2001 November 19 18:25:51 UT, duration 4.88  ks) was examined. No sources were found by wavdetect within 70'' of the RASS/BSC position. The Chandra observation only reveals the highly resolved M31 galaxy, with no point source in the vicinity of the RASS/BSC position.

We note the presence of an X-ray source CXO J004337.27+411443.4 approximately 75'' from the RASS/BSC position, with a count rate of 45.7±3.7 counts ks⁻¹, which corresponds to a flux of 6.3×10⁻¹³ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The total positional uncertainty of the RASS/BSC source is ± 26'', placing this object 2.9σ from the RASS/BSC position. This Chandra source is at the same position as an XMM detected X-ray source XMMM31 J004337.2+411444, which is associated with a globular cluster at the distance of M31; this object has, at the 770 kpc distance of M31, a luminosity of 1.7×10³⁷ erg s⁻¹. The observed modestly lower (×3.2) count rate is consistent with the spectral uncertainty and possible intrinsic source variability which may be associated with a LMXB in a globular cluster. The probability of chance association between the Chandra/ACIS-I source and a background AGN is P_chance = 10^−2.3.

We therefore conclude that 1RXS J4330.9+411452 is the X-ray source CXO J004337.27+411443.4, and the X-ray source XMMM31 J004337.2+41144; not an INS.

1RXS J004704.8-204743. The galaxy NGC 247 was observed with ROSAT/HRI (Lira et al. 2000). Two bright X-ray sources were found south of the galaxy (denoted X1 and X2 in Lira et al. 2000). By visual inspection, the RASS/BSC source appears spatially coincident with X1 (and inconsistent with X2, which lies ∼1' away). X1 is found to have a very soft spectral distribution (Read et al. 1997), with a bremsstrahlung spectrum with kT = 120⁺³⁰_-20 eV and N_H,22 = 0.6^+0.1_−0.2. A factor of 2 variability in X-ray flux over a timescale of years is also found (Lira et al. 2000). While previous observers conclude this object is a ULX in NGC 247 based on its factor of 2 variability and consistency with a high luminosity (10³⁹ erg s⁻¹) X-ray source at the distance of NGC 247, we cannot exclude that it is a foreground INS due to the lack of an identified off-band counterpart; undetermined.

1RXS J020317.5−243832. Not an INS (R03).

1RXS J024528.9+262039. Not an INS (R03).

1RXS J025414.5+413530. This RASS/BSC source has a high count rate (2.64 ± 0.05 PSPC counts s⁻¹), and is extended on length scales of ∼15'. In a 50  ks Chandra observation (observation ID = 908), an X-ray source extended on length scales of ∼5' is located ∼25 from the RASS/BSC position. The next nearest RASS/BSC source is ∼42' away from the extended source. The extended source has been identified as the galaxy cluster AWM 7 (Furusho et al. 2003). We suggest therefore that this RASS/BSC source is AWM 7, and that the RASS/BSC astrometry for this object is incorrect by 26; not an INS.

1RXS J034414.1+240623. This X-ray source was redetected in 30 ks of ROSAT/HRI observations (Panzera et al. 2003), with a position 03^h44^m148, +24°06'057, and an HRI count rate of 20±1 counts ks⁻¹, consistent with the detected RASS/BSC PSPC count rate (64.5 ± 15.2 PSPC counts ks⁻¹; 24.3 counts ks⁻¹ expected for HRI). The probability of chance association between the ROSAT/HRI source and a background AGN is P_chance = 10^4.0. We therefore conclude that the HRI source is associated with the RASS/BSC source.

The closest cataloged USNO-B1 optical source is USNO-B1 1141-0042520 (R = 9.99), located 38 away from the HRI source position. Within a radius of 1000'' we find eight objects with as bright or brighter R-band magnitudes, producing a probability of chance spatial coincidence of ∼10⁻⁴. We therefore associate the RASS/BSC X-ray source with this optical source, which is identified as BD+23 502; not an INS.

1RXS J051541.7+010528. This X-ray source has been observed with XMM-Newton  and found to have an optical counterpart, identified as the long-period polar V1309 Orionis; not an INS (Schwarz et al. 2005).

1RXS J053510.8−044850. Observed with Chandra/ACIS-I (observation ID = 2549, observation start time: 2002 August 26 13:49 UT, duration 48.8  ks).

Based on the RASS/BSC count rate (200 ± 25 PSPC counts ks⁻¹), the expected Chandra/ACIS-I count rate is 164 counts ks⁻¹ which corresponds to a total of ∼7940 counts detected during the Chandra observation from this source.

No individual point source with this count rate is found within 90'' of the RASS/BSC position (which is uncertain by 26'', 1σ). We calculate a 99% confidence upper limit to the flux (i.e., count rate) of this X-ray source during the Chandra observation, assuming a point source, of <0.29 counts ks⁻¹, which corresponds to a flux of <4.3×10⁻¹⁵  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV), which is approximately a factor of 2 × 10³ fainter than the RASS/BSC measured flux of 8.4×10⁻¹²(0.1–2.4 keV).

There is a faint X-ray source located approximately 30'' from the RASS/BSC position, with a count rate of 0.4 ± 0.1 counts ks⁻¹ (flux of ∼5.9×10⁻¹⁵  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹). The probability of chance association between the Chandra/ACIS-I source and a background AGN is P_chance = 0.35.

This RASS/BSC source is near an optical nebulosity in the DSS. A wavdetect imaging analysis was performed with the Chandra data, searching for sources on spatial scales of 2'', 4'', 8'', 16'', 32'', 64'', and 128''; no spatially resolved X-ray source was found within 60'' of the RASS/BSC position at the flux level detected from the RASS/BSC source. Thus, the nature of this RASS/BSC source cannot be determined; undetermined.

1RXS J053516.3−044033. Observed with Chandra/ACIS-I (observation ID = 2549, observation start time: 2002 August 26 13:49 UT, duration 48.8 ks). A wavdetect analysis finds a point source located at 05^h35^m412, −04°40'31'' (±006 statistical, ±06 systematic). This source was found to have a count rate of 39.5 ± 0.9 counts ks⁻¹, which corresponds to a flux of 5.8×10⁻¹³  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The RASS/BSC count rate (52.4 ± 13.4 PSPC counts ks⁻¹) corresponds to a flux to be 2.2 ×10⁻12  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). Thus, the Chandra source is found to be a factor of ×4 fainter than the RASS/BSC source. The probability of chance association between the Chandra/ACIS-I source and a background AGN is P_chance = 10^−4.8. We therefore conclude that the Chandra X-ray source is the RASS/BSC source, now fainter by a factor of ∼7.

2MASS J053516.7−044032 is located 095 away from the Chandra source position. The probability of chance association is P_chance = 10^−2.9; not an INS.

1RXS J053842.4−023525. Observed with Chandra/HRC-I (observation ID = 2560, duration 98.4 ks). A wavedetect analysis detects two X-ray sources, 3'' apart, with positions 05^h38^m447, −02°36'00'' (±0001 statistical, ±06 systematic) and 05^h38^m448, −02°35'57'' (±001 statistical, ±06 systematic). This binary is located 49'' from the RASS/BSC source position (which has an uncertainty of ± 11''). Their position is 02 away from the binary Sigma Orionis AB, USNO J053844.7−023600. The probability of chance association is P_chance = 10^−2.9, and we conclude the Chandra/HRI source is the binary Sigma Orionis AB.

The RASS/BSC count rate (638±40 PSPC counts ks⁻¹) corresponds to a flux of 2.6×10⁻¹¹ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). We determine the Chandra count rate, for both sources added together, to be 809 ± 2 counts ks⁻¹, which corresponds to a flux of 2.1×10⁻¹¹  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the Chandra/HRC-I source and a background AGN is P_chance = 10^−4.7. We conclude the Chandra/HRC-I X-ray sources to be 1RXS J053842.4−023525; not an INS.

1RXS J054042.8−020533. During 42 ks of observations with XMM-Newton/MOS-1 (observation ID = 0101440301), an X-ray source is detected with a position 05^h40^m413, = 02°05'38'' (±05 statistical, ±2'' systematic), 23'' away from the RASS/BSC position. It was found to have a count rate of 6.4 ± 0.5 counts ks⁻¹, which corresponds to a flux of 1.3 ×10⁻¹³ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The RASS/BSC source has a count rate of 177 ± 22.1 counts ks⁻¹, which corresponds to a flux of 7.5×10⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the XMM-Newton source and a background AGN is P_chance = 10^−1.5.

We note the presence of a source located at 05^h40^m540, −02°03'02'', ∼38 from the RASS/BSC position. This source has a count rate of 72.0 ± 1.3 counts ks⁻¹, which corresponds to a flux of 1.5×10⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the XMM-Newton source and a background AGN is P_chance = 10^−1.8.

Thus, we cannot conclude that either of the two XMM-Newton X-ray sources are associated with the RASS/BSC source. Undetermined.

1RXS J054045.7−021119. A wavdetect analysis was performed on a 7.4  ks duration ROSAT/HRI observation (observation ID = rh201394n00). An X-ray source was localized at 05^h40^m447, −02°11'54'', (± 14, statistical, ± 6'' systematic), 37'' from the RASS/BSC source position (which has a ± 18'' uncertainty).

The RASS/BSC source count rate (161 ± 20 PSPC counts ks⁻¹) corresponds to a flux of 6.8×10⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹ (0.5–2.0 keV). The HRI source was found to have a count rate of 4.47 ± 0.81 counts ks⁻¹, which corresponds to a flux of 5.0×10⁻¹³ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the ROSAT/HRI source and a background AGN is P_chance = 10^−2.8. We therefore conclude that the HRI source is associated with the RASS/BSC source.

A B = 13.48 USNO-B1.0 source is located 46 away from the HRI source position, at 05^h40^m44.82^s, −02°11'500. For objects as bright or brighter than B = 13.48, the probability of chance association is 10^−3.0. We conclude that the HRI object is the RASS/BSC source, which has faded by a factor of ∼7; not an INS.

1RXS J055228.1+155313. The ROSAT/PSPC count rate is 2.14 ± 0.07 counts s⁻¹. No higher spatial resolution observations (ROSAT/HRI, Chandra, or XMM-Newton) exist of this source. The nearest USNO-B1 object (USNO-B1 1058-0090877) lies 69 away, and is a spectroscopically identified WD GD 71, which has been previously associated with the RASS/BSC source (Fleming et al. 1996). Based on the R-band magnitude (R = 12.96), and that there are 164 USNO-B1 objects as bright or brighter within 1000'', the probability of chance association is 7×10⁻³; not an INS.

1RXS J055800.7+535358. This object has been associated with an optical counterpart that is a known intermediate polar, with an X-ray period of 272.74 s X-ray (Haberl et al. 1994); not an INS.

1RXS J060452.1−343331. During 21.3 ks of observations with the ROSAT/HRI (observation ID = rh202301n00), an X-ray source is detected with a position 06^h04^m5211, −34°33'3494. (±01 statistical, ±6'' systematic).

The RASS/BSC source has a count rate of 428 ± 50 PSPC counts ks⁻¹, which corresponds to a flux of 8.8×10⁻¹²  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The ROSAT/HRI source was determined to have a count rate of 86.2 ± 2.0 counts ks⁻¹, which corresponds to a flux of 5.3×10⁻¹²  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the ROSAT/HRI source and a background AGN is P_chance = 10^−4.5.

A B = 13.81 USNO-B1.0 source is located 12 away from the HRI source position at 06^h04^m5215, −34°33'360. The probability of chance association is 10^−3.3. This optical star is an eruptive variable V* AP Col, which has previously been associated with the 1RXS J060452.1−343331 (Fuhrmeister & Schmitt 2003); not an INS.

1RXS J063354.1+174612. This X-ray source is associated with the radio-quiet pulsar SN 437 (Geminga). It was associated based on its X-ray pulsations, which have a period of 0.237 s. (Halpern 1992). For purposes of this paper, we treat this object as an INS.

1RXS J074451.8+392733. This X-ray source has been observed with Chandra and found to be significantly extended, and spatially associated with a galaxy cluster (Ebeling et al. 2007); not an INS.

1RXS J085247.0+223040. Performing a wavdetect analysis on XMM-Newton/MOS-1 observation ID = 0144500101 (0.7 ks duration), an X-ray source was localized at 08^h52^m446,  +22°30'52'' (±100 statistical, ±2'' systematic).

The XMM-Newton source was found to have a count rate of 71 ± 17 counts ks⁻¹, which corresponds to a flux of 8.6×10⁻¹³ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The RASS/BSC source count rate is 165 ± 25 PSPC counts ks⁻¹, or 3.2×10⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the XMM-Newton source and a background AGN is P_chance = 10^−2.8.

A USNO-B1.0 source, USNO J085244.7+223054, is located 22 away from the XMM-Newton position. For objects as bright or brighter than B = 14.61, the probability of chance association is 10^−3.9; not an INS.

1RXS J091112.2+174634. This X-ray source has been observed with Chandra and found to be significantly extended, and spatially associated with a cluster of galaxies (Ebeling et al. 2007); not an INS.

1RXS J101628.3−052026. Previously identified with a spectroscopically classified WD (Fleming et al. 1996), at 10^h16^m286, −05°20'27'', with B = 13.3, V = 14.3, R = 12.02 (USNO-B1.0 0846-0208321), and an approximately co-local 2MASS source 2MASS J101628.67−052032.0 with (J = 10.607, H = 9.99, K = 9.77). The 2MASS source is 7'' from the RASS/BSC position, and there are seven such objects as bright or brighter as J = 10.607 within 1000'', making the probability of chance positional coincidence ∼3×10⁻⁴; not an INS.

1RXS J102954.3+614732. This object was observed with ROSAT/HRI (observation ID = rh704085n00, 4.9  ks). Two point sources lie 1' and 15 from the RASS/BSC position, with no other detected X-ray sources within 3'. The position of the closer HRI X-ray source is 10^h29^m50208, +61°46'4565 (±1'' statistical, ±6'' systematic).

The RASS/BSC source has a count rate of 145 ± 18 PSPC counts ks⁻¹, which corresponds to a flux of 1.4×10⁻¹²  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV), assuming N_H = 7.8×10¹⁹ cm⁻² and a photon power-law slope of α = 2. The HRI source was determined to have a count rate of 18.3 ±1.8 counts ks⁻¹, which corresponds to a flux of 7.6×10⁻¹³.  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the ROSAT/HRI source and a background AGN is P_chance = 10^−2.6.

The factor of ∼×2 difference in fluxes is considered to be consistent with spectral uncertainties and source variability, and we therefore associate the HRI X-ray source with the RASS/BSC X-ray source.

The optical source USNO J102950.97+614643.2 (B = 17.35) is located 59 away from the HRI source position. The probability of chance association with an object as bright or brighter than B = 17.35 is 10^−2.3; not an INS.

1RXS J103347.4−114146. There have been no X-ray observations with the specified instrumentation; undetermined.

1RXS J104710.3+633522. Source has been previously found (R03) to be a magnetic cataclysmic variable (Singh et al. 1995); not an INS.

1RXS J115309.7+545636. This object was found to exhibit variability in the X-ray band, being undetected at a flux limit lower than that of the RASS/BSC detection (R03); undetermined.

1RXS J123319.0+090110. This X-ray source has been found to be a DM star (R03; Marino et al. 2000); not an INS.

1RXS J130547.2+641252. Observed with HRI but undetected; therefore, excluded previously as an INS on the basis of X-ray variability (R03). However, in the present analysis, X-ray variability is not grounds for exclusion as an INS candidate.

XMM-Newton/PN observation ID = 0151790701 (6.9  ks duration) was examined. Performing a wavedetect analysis revealed that the X-ray point source closest to the RASS/BSC position is 175'' away, at 13^h06^m016, +64°10'27''. The RASS/BSC X-ray source positional uncertainty is ± 9''.

The RASS/BSC count rate (167 ± 20 PSPC counts ks⁻¹) corresponds to a flux of 2.3×10⁻¹²  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The XMM-Newton source was determined to have a count rate of 15.6 ± 1.6 counts ks⁻¹, which corresponds to a flux of 4.2×10⁻¹⁴  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the XMM-Newton source and a background AGN is P_chance = 0.45. We cannot conclude that the two sources are related.

We calculate a 99% confidence upper limit to the flux of this X-ray source during the XMM-Newton observation, assuming a point source of <2.0 counts ks⁻¹, which corresponds to a flux of <5.3 ×10⁻¹⁵ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV), which is a approximately a factor of 400× fainter than the RASS/BSC measured flux of 2.3×10⁻¹²(0.1–2.4 keV); undetermined.

1RXS J130753.6+535137. Not an INS (R03).

1RXS J130848.6+212708. Previously identified INS (Hambaryan et al. 2002).

1RXS J132833.1−365425. Not an INS (R03).

1RXS J134210.2+282250. Identified as a CV in M3 (R03; Dotani et al. 1999); not an INS.

1RXS J145010.6+655944. Not an INS (R03).

1RXS J145234.9+323536. This object was previously found to found to exhibit variability in the X-ray band (R03), which in the present analysis is no longer grounds for exclusion as an INS candidate. Observed with Chandra, but no X-ray source was detected (R03); undetermined.

1RXS J160518.8+324907. Previously identified INS (Motch et al. 1999); INS.

1RXS J161455.4−252800. Performing a wavdetect analysis on ROSAT/HRI observation (observation ID = rh201993n00, duration 3.25 ks), we find a source 24 away from the RASS/BSC position, at 16^h14^m553, −25°27'57'' (±08 statistical, ±6'' systematic). This HRI source was found to have a count rate of 37 ± 4 counts ks⁻¹, which corresponds to a flux of 3.8×10⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The RASS/BSC count rate is 100 ± 20 counts ks⁻¹, which corresponds to a flux of 3.9×10⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the ROSAT/HRI source and a background AGN is P_chance = 10^−6.3. We conclude that the two sources are associated.

USNO J161455.4−252800 lies 32 away from the HRI position. For objects as bright or brighter than R = 9.33, the probability of chance association is 10^−4.2; not an INS.

1RXS J162721.6−244144. Observations using Chandra/ACIS-I of the rho Ophiuchi region (Imanishi et al. 2001) localize the nearest X-ray source to 16^h27^m214, −24°41'43'', (±09, 90% confidence). The RASS/BSC count rate (93.7±17.8 PSPC counts ks⁻¹) corresponds to an expected Chandra ACIS-I count rate of 137 counts ks⁻¹. Chandra observation photon count rate for this source is 1.39 counts ks⁻¹, a factor of 100× fainter than the expected count rate.

There is a much brighter Chandra X-ray source 29'' away from the RASS/BSC position (which has ± 14'' uncertainty), localized at 16^h27^m195, −24°41'40''. This source was found to have a count rate of 83.8 counts ks⁻¹(Imanishi et al. 2001), a factor of 0.6× the expected ACIS-I count rate. The probability of chance association between the Chandra/ACIS-I source and a background AGN is P_chance = 10^−3.5. We conclude that the two sources are associated.

There is a nearby USNO source, USNO J162719.5−244140, 04 away from the position of the brighter Chandra source. The probability of chance association is P_chance = 10^−3.3. Thus, we conclude that 1RXS J162721.6−244144 = CXO J162719.5−244140.6 = USNO J162719.5−244140; not an INS.

1RXS J163212.8−244013. A wavdetect analysis was performed with ROSAT/HRI observation (observation ID = rh201839n00, duration 6.33  ks). An X-ray source was localized at 16^h32^m119, −24°40'21'' (± 13 statistical, ± 6'' systematic), 14'' from the RASS/BSC source position (which has uncertainty of ± 10'').

Based on the RASS/BSC count rate (78.6 ± 15.8 PSPC counts ks⁻¹), the RASS/BSC source has a flux of 2.8×10⁻¹²  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV) The HRI source was found to have a count rate of 24.6±2.0 counts ks⁻¹, which corresponds to a flux of 2.3×10⁻¹²  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). The probability of chance association between the ROSAT/HRI source and a background AGN is P_chance = 10^−4.5.

A B = 15.27 USNO-B1.0 source, USNO J163211.8−244021, is located 17 from HRI position. For objects as bright or brighter than B = 15.27, the probability of chance association is 10^−3.6; not an INS.

1RXS J163421.2+570933. Not an INS (R03).

1RXS J163910.7+565637. Not an INS (R03).

1RXS J173157.7−335007. This source has been identified with a known and well-studied LMXB, MXB 1728-34 (Liu et al. 2001); not an INS.

1RXS J173253.6−371200. - This object has no X-ray observations with the specified instrumentation; undetermined.

1RXS J173319.3−255416. There is a RASS/BSC source 15'away (by visual inspection), 1RXS J173413.0−260527, also known as KS 1731-261, a cataloged LMXB. By visual inspection of the RASS image, there does not appear to be any independent X-ray source at the given position for 1RXS J173319.3−255416, which is in the PSF wing of KS 1731-261.

Over 200 ks of integration with XMM-Newton have been accumulated in which the RASS/BSC position is 16' off-axis, and 90'' outside the field of view (the observations target KS 1731-261).

We then examined two ROSAT/HRI observations of the LMXB, observation ID = rh400718n00 (1.08  ks) and observation ID = rh400718a01 (4.68  ks). The RASS/BSC count rate (59.2 ± 17.2 counts ks⁻¹) corresponds to a flux of 8.7×10⁻¹³  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). Thus, the expected ROSAT/HRI count rate is 222 counts ks⁻¹.

For both observations, we find no such bright individual point sources within 90'' of the RASS/BSC position (which is uncertain by 14'' , 1σ).

For observation ID = rh400718n00, we calculate a 99% confidence upper limit to the flux of this X-ray source during the ROSAT/HRI observation, assuming a point source of <10.2 counts ks⁻¹, which corresponds to a flux of <3.98×10⁻¹³  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV), which is approximately a factor of 2× fainter than the RASS/BSC measured flux.

Our conclusion is that this RASS/BSC catalog entry is due to a spurious source detection in the wing of KS 1731-260; not an INS.

1RXS J173546.9−302859. This X-ray source has been associated with a neutron star X-ray transient X1732-304 in quiescence in the globular cluster Terzan 1 (Wijnands et al. 2002); not an INS.

1RXS J180132.3−203132. This X-ray source has been associated with a well-studied LMXB, X SGR X-3 (4U 1758−20), including recent simultaneous X-ray optical observations (Kong et al. 2006); not an INS.

1RXS J181506.1−120545. This X-ray source has been associated with a faint LMXB (XB 1812-12). A Chandra observation reveals an object which is 24 offset from the RASS/BSC position, (18^h15^m061,−12°05'47'') with a flux of 4.4 × 10⁻¹⁰ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(1–10 keV) (Wilson et al. 2003), which is comparable to the flux of the RASS/BSC source (1.3 ± 0.1 PSPC counts s⁻¹, corresponding to 9.3 × 10⁻¹¹  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹). The probability of chance association between the Chandra source and a background AGN is P_chance = 10^−9.2. We therefore conclude that the two objects are associated; not an INS.

1RXS J182102.0−161309. No optical counterpart was found in USNO-A2, although the DSS image was found to contain a nebulosity (Rutledge et al. 2000).

This X-ray source has been observed with XMM-Newton (observation ID = 0144500101, observation date 2003 March 11, duration 35 ks, 8.9 ks after flare removal), approximately 74 off-axis.

The RASS/BSC count rate (161 ± 28 PSPC counts ks⁻¹) corresponds to a flux of 2.9×10⁻¹¹  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). Thus, the expected XMM-Newton count rate is 1440 counts ks⁻¹, which corresponds to a total of ∼12,816 counts detected during the XMM-Newton observation of this source.

A wavdetect analysis, searching for sources on spatial scales of 4'', 8'', 16'', 32'', 64'', and 128'' of the PN image localized one X-ray source located 150'' from the RASS/BSC position (which has an uncertainty of ± 37'').

The brightest nearby source was found to have a count rate of 3.9 ± 1.1 counts ks⁻¹ which corresponds to a flux of 3.3 × 10⁻¹⁴ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.5–2 keV). The probability of chance association between the XMM-Newton source and a background AGN is P_chance = 0.10. This does not support a conclusion that the brightest XMM-Newton source is associated with the RASS/BSC source.

We calculate a 99% confidence upper-limit to the flux of this X-ray source during the XMM-Newton observation, assuming a point source of <1.69 counts ks⁻¹, which corresponds to a flux of <3.4 ×10⁻¹⁴  erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV), which is approximately a factor of ×800 fainter than the RASS/BSC measured flux; undetermined.

1RXS J183543.6−325928. This X-ray source has been associated with an LMXB (XB 1832-330) in the globular cluster NGC 6652. (Liu et al. 2001); not an INS.

1RXS J185557.3+233410. This object has no X-ray observations with the specified instrumentation; undetermined.

1RXS J191426.1+245641. This object has been associated with a cataclysmic variable V* V407 Vul (Ramsay et al. 2000), with extensive Chandra and XMM-Newton observational history; not an INS.

1RXS J221144.6−034947. In an 80 ks Chandra observation (observation ID = 3284), an X-ray source is localized using wavdetect, at 22^h11^m458, −03°49'48'', with ± 015 statistical, ± 06 systematic uncertainty, located ∼ 18'' from the RASS/BSC position. The wavdetect analysis indicates that this source has a PSF ratio of 5.81, so we consider it to be extended. The RASS/BSC source has been identified as galaxy cluster RXC J2211.7−0350 (Böhringer et al. 2004); not an INS.

1RXS J221403.0+124207. This X-ray source has been associated with a historical dwarf nova RU Peg (Silber et al. 1994). While no X-ray observations with the specified instrumentation have been analyzed, one such observation with XMM-Newton which was taken in 2008 June 9 can be used to confirm the localization; we tentatively conclude this X-ray source is the dwarf nova RU Peg; not an INS.

1RXS J223832.0−151809. A wavedetect analysis on a pointed ROSAT/PSPC observation (observation ID = RP201723N00, 13.2 ks duration) determines the X-ray position to be 22^h38^m32040, −15°18'0391 (±04 statistical, 6'' systematic). Archival images in DSS and 2MASS (1982, 1987, 1991, and 2000) reveal a high proper-motion star which was located within the X-ray error circle at the time of the RASS/PSPC observation. 2MASS J223833.7−151757 (J = 6.6) is located 25'' away from the PSPC source position. The probability of chance association for this object is 10^−3.7; not an INS.

1RXS J230334.0+152019. We analyzed a ROSAT/HRI observation (observation ID = rh701842n00, 7.3  ks duration). A wavdetect imaging analysis was performed, searching for sources on spatial scales of 2'', 4'', 8'', 16'', 32'', and 64''; a source was found 24'' away from the RASS/BSC position, at 23^h03^m323, +15°20'16'' (± 05 statistical, ± 6'' systematic).

The RASS/BSC count rate (60.2 ± 13.7 PSPC counts ks⁻¹) corresponds to a flux of 1.5×10⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). We determine the ROSAT/HRI count rate to be 9.6 ± 1.2 counts ks⁻¹, corresponding to a flux of 6.9×10⁻¹³ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹ (0.1–2.4 keV), a factor of ×2 fainter than the RASS/BSC flux. The probability of chance association between the ROSAT/HRI source and a background AGN is P_chance = 10^−3.3. We therefore conclude that the HRI source is the RASS/BSC source, now fainter by a factor of ∼2.

The closest cataloged USNO-B1 optical source is USNO J230332.4+152009 (B = 20.4), located 63 away from the HRI position. For objects as bright or brighter than B = 20.4, the probability of chance association is 10^−1.4; undetermined.

1RXS J233407.2−033533. There have been no X-ray observations with the specified instrumentation; undetermined.

1RXS J233757.2+271031. We examined the XMM-Newton/MOS-1 observation (observation ID = 0002960101, 11.5  ks). A wavdetect imaging analysis was performed, searching for sources on spatial scales of 2'', 4'', 8'', 16'', 32'', 64'', and 128''; a source was found 57'' away from the RASS/BSC position, at 23^h37^m548, +27°11'10''(± 09 statistical, ± 2'' systematic).

The RASS/BSC count rate (68.4±12.9 PSPC counts ks⁻¹) corresponds to a flux of 1.710⁻¹² erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV). We find the XMM-Newton/MOS-1 source count rate to be 34.6 ± 2.1 counts ks⁻¹, which corresponds to a flux of 4.5×10⁻¹³ erg $\mbox{$\rm \,cm$}^{-2}$ s⁻¹(0.1–2.4 keV), a factor of ×4 fainter than the RASS/BSC flux. The probability of chance association between the XMM-Newton source and a background AGN is P_chance = 10^−2.3. We therefore conclude that the XMM-Newton X-ray source is the RASS/BSC source, now fainter by a factor of ∼4.

The XMM-Newton source is located 95 away from USNO J233756.0+271117 (B = 18.46). The probability of chance association is P_chance = 10^−1.4. The XMM-Newton source is also located 111 away from 2MASS J233755.5+271113 (J = 15.7). The probability of chance association is P_chance = 10^-1.3; undetermined.