Tables 1 and 2 are only available in electronic form at the CDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or via http://cdsweb.u-strasbg.fr/cgi-bin/qcat?J/A+A/429/963

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

Appendices are only available in electronic form at http://www.edpsciences.org

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... statistics

There is a coding error in emldetect until SAS version 5.4 (XMM-Newton-News #29, 11-Mar.-2003), producing an overestimate of the source likelihood. We corrected the source likelihood using the table provided in the XMM-Newton-Newsletter.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...'')

The official naming convention for XMM-Newton sources is XMMU JHHMMSS.s+/-DDMMSS.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... respectively

ROXN-1 was so bright that the CCD pixels on the source are strongly affected by pile-up. Hence HR computed for this source is not reliable.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... 1.5 keV

See calibration files XRT1 XPSF 0005.CCF, XRT2 XPSF 0005.CCF, and XRT3 XPSF 0004.CCF.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... curves

For ROXN-1, we exclude a circular region of 17.5 $^{\prime \prime }$ radius centered on the source to keep the spectrum from being affected by pile-up.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... (WABS)

We check the consistency of $N_{\rm H}$ determinations between Chandra and XMM-Newton in Appendix B. We mention also the uncertainties of X-ray derived $N_{\rm H}$ values especially due to a change in metal abundances for the absorption model.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... survey

Note that the prototype of Class 0 sources, VLA1623 (André et al. 1993) is located in the dense core A, which is not covered by our observation.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... dwarfs

We note in Fig. 8 that ROXN-82 is located well below the substellar limit of $\sim$ 0.08 $M_{\odot }$ and could be a brown dwarf candidate, but it was found to be a very low-mass M star by a near-IR spectroscopical follow-up (Grosso et al. 2004, in preparation).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... archive

http://asc.harvard.edu/cgi-gen/cda/retrieve5.pl

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.