Near--infrared observations were made on 5--6 February 1996 at CTIO using the CIRIM imager. CIRIM is a 256256 HgCdTe array sensitive to near infrared wavelengths. At the focus of the CTIO 1.5 meter, the pixels are 1.16 on a side. The field of view was about 5, and J, H and K filters were used. Sixteen fields near Orionis were observed. Exposure times averaged about 10 seconds in each filter with six coadds per exposure. In the infrared, the sky background is high relative to the dynamic range of the detector. This effect is mild at J--Band, but it limits K--band exposures to under 10 seconds. It is desirable to coadd several exposures together. The nominal seeing at the CTIO 1.5 meter was about 0.8 arcseconds. Since the pixel size was larger than the nominal size of a seeing disk, it was necessary to move the telescope several times during the exposure sequence to get an accurate flux measurement. This process is called dithering. Five dithered exposures were taken of each field. As the first step in the reduction, all frames had a linearity correction applied (Elston 1994). Afterward, the individual frames were dark--subtracted and flatfielded. The dithered frames, thus reduced, were then added together to create a composite frame and median filtered to create a sky background frame. This sky background frame was multiplied by five and then subtracted from the composite frame to produce a final frame.
Photometry was carried out only on those sources for which a possible (not necessarily confirmed) optical counterpart to an X--ray source had been identified. The IRAF/APPHOT package was used in a similar manner to the optical data with a similar 4" radius aperture. In some cases, the dithering procedure caused stars to overlap with other stars on the final frame. In these cases, corrected single star frames were measured using IRAF/APPHOT. Comparisons stars were taken from the UKIRT faint standards (Casali & Hawarden 1992). Standard stars were observed between five and nine times at each airmass; airmasses ranged from 1.2 to 2.2. The solution for each night was accurate to about 5% in each color. These analyses lead to five measurements of each target star in each filter. The results were averaged and are listed in Table 8. The errors given are the standard deviations among the five measurements.