... stars[*]
Tables 1, 3, 4 and 6 are only available in electronic form at http://www.edpsciences.org 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... Center[*]
See cxc.harvard.edu 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... sources[*]
${\it SNR}_{\rm tot}=\sum{(S^i- \alpha_i B)}/\sqrt{\sum{(S^i+ \alpha_i B)}}$ where $\alpha_i=A^i_s/A_b$; Si and Ais are the number of counts in the ith source region and its area, respectively. B and Ab are the background counts and relative extraction area, respectively. The background was taken for the present purpose as uniform within the FOV. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...(Damiani et al. 1997)[*]
Available at: http://www.astropa.unipa.it/progetti_ricerca/PWDetect 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... keV[*]
The choice of energy is not crucial: the ratio between the effective area at the source position and on the optical axis, which is the quantity we use to define the effective exposure time, has only a small dependence on energy. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... spectroscopy#[*]
The Rebull et al. (2002), catalog was updated following private communication from L. Rebull. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... (JHKs[*]
Note that in the following we neglect the small differences among NIR photometric systems and, in particular, that between the 2MASS Ks and the standard K bands. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... track[*]
This latter condition was required by the convergence of the isochrones at higher masses and by the observed spread of stars in the "cluster locus'', at least in part due to uncertainties. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... Y06)[*]
Note that the source is missed by 2MASS because, at the 2MASS spatial resolution, it is blended with a nearby source. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...$EW({\rm H}_\alpha)=231$)[*]
Although Y06 report that C-MM 1 has no NIR counterpart, we associate it with 2MASS source 06411792+0929011 (offset = 1.5''). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... interval[*]
Events with energies below 0.5 keV were not used because of uncertainties in the calibration resulting from the time-dependent degradation of the ACIS quantum efficiency at low energies. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... acceptable[*]
For source #241 we changed the 2T model ( $\rm n.p.=32$%, kT1=0.0577.80.022 keV, kT2=0.571.300.30 keV, $N_{\rm H}=13^{21}_{1.8}\times 10^{21}\ \rm cm^{-2}$) to a 1T model ( $\rm n.p.=17$%, kT=0.600.840.38 keV, $N_{\rm H}=3.3^{6.5}_{0.0}\times 10^{21}\ \rm cm^{-2}$). For source #104 we changed the 1T model ( $\rm n.p.=40$%, kT=0.430.730.29 keV, $N_{\rm H}=3.0^{5.6}_{0.78}\times 10^{21}\ \rm cm^{-2}$) to a $2{\rm T}_{A{\rm v}}$ model (n.p. = 32%, $N_{\rm H}(A_{\rm V})=0.0~\rm cm^{-2}$ and reasonable temperatures). For sources #97 and #127 we changed from $1{\rm T}_{A{\rm v}}$ models (p=68% and p=21%, kT=54 keV and kT=19 keV, both with unconstrained uncertainties) to 1T models (p=97% and p=56%, kT=4.7 keV and kT=3.5 keV). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... cases[*]
Sources #190, #340, #352, #375, #391. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...$N_{\rm H}=1.1\times 10^{21}$)[*]
For this source we adopted the spectral type (K2) from Dahm & Simon (2005) and the photometry ( $R_{\rm c}-I_{\rm c}=0.62$) from Lamm et al. (2004). Independent photometry and spectral types are available from Rebull et al. (2002), Lamm et al. (2004), and Dahm & Simon (2005). The $R_{\rm c}-I_{\rm c}$ color ranges between 0.55 and 0.62 and spectral types between G9 and K2. The possible range in $N_{\rm H}=1.6\times 10^{21}A_{\rm V}$ is thus $0.57{-}1.64\times 10^{21}$, i.e. between $\sim$9 and $\sim$46 times the value derived from the X-ray spectrum, considering its 90% confidence interval. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...)[*]
We excluded 14 sources identified as stars by Barger et al. (2003). 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... objects[*]
Mean value and 1$\sigma $ uncertainty result from repeating the experiment 10 times, each time varying the position of simulated extragalactic sources. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... diagrams[*]
The NIR photometry of the CDFN sources was taken from 2Mass. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...(Schreyer et al. 2003)[*]
The ACIS spectrum indicates $N_{\rm H}=4.8^{8.9}_{3.2}\times 10^{22}$ (90% confidence interval), corresponding to $A_{\rm V}= 20{-}56$. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... stars[*]
#7, #51, #110, #128, #130, #171, #224, #248, #258, and #267. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... neighborhood[*]
This is justified by the low equatorial longitude of NGC 2264: $2.2\deg$. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... cm[*]
With the chosen value of $\rho$, taking $N_{\rm H}/A_{\rm V}=1.6\times 10^{21}$ and d=760 pc (the assumed distance of NGC 2264), we obtain $A_{\rm V}=0.44$, in agreement with the median of values measured for NGC 2264 members. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... non-members[*]
The fraction of non-members is expected to be higher among X-ray sources with $M/M_\odot > 1$ and ages ${\sim} 10^7$ yr. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... considered[*]
Four of these X-ray sources were actually identified with MIR and/or mm sources in Sect. 3.3. Because these catalogs are not spatially complete, we here treat them as unidentified for uniformity. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... HRI[*]
In Table 3 our ACIS source # 376 is associated with HRI source 138 due to the large uncertainty of the HRI positions; the HRI source is, however, closer to the brighter ACIS source #375, with which it is most likely associated. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...$\sim$1.9[*]
80% interval of the power indexes reported by Alexander et al. (2003) for the sources in the CDFN that we would have detected in our ACIS data and for which the 1$\sigma $ uncertainty in the index is lower than 0.1. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... ONC[*]
Note that the COUP observation, obtained with Chandra ACIS, was 850 ks long. The temperatures are thus derived from a spectrum that has been integrated in time over a period that is 8.5 times longer than that of our NGC 2264 observation. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...spectrum[*]
Our $N_{\rm H}$ is based on $A_{\rm V}=0.58$, derived here from published photometry and spectral type. Herbst & Moran (2005) derive their value from E(B-V)=0.1. This E(B-V) should, however, imply $N_{\rm H}=5{-}6\times 10^{20}$ cm-2 assuming $A_{\rm V}/E(B-V)=3.1$ and depending on the assumed $A_{\rm V}/N_{\rm H}$ ratio in the range $1.6{-}2.0 \times 10^{21}$. Herbst & Moran (2005) assume solar abundances and kT=2.7 keV, the latter based on the uncertain (due to statistics) hardness ratio. The count to flux conversion factor we used (Eq. (1)) was derived from sources with 50 to 100 counts, which were fit with APEC models with subsolar abundances and median kT of 1.0 keV. Note that our harness ratio analysis (cf. Fig. 16) actually indicates a very cool spectrum, $kT\sim 0.3$ keV, with large uncertainties. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
... uncertainties[*]
In this section uncertainties quoted for quantities derived by SD05 are $1\sigma$, while for our results we quote 90% confidence intervals. 
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Copyright ESO 2006