F. Walter (1996) obtained WIYN/HYDRA queue observations of stars in both
the 
Orionis field and the field NW of the belt.
The targets fell in two categories:
A. Stars which are likely counterparts of the X--ray sources.
These were identified by visual inspection of optical images.
B. Stars with magnitudes and colors comparable to those selected
as likely X--ray counterparts.
Observations were in the 5850--6800Å band at 1Å resolution (Grating
860). The integration time was an hour on each field. This yielded S/N of
about 100 per pixel in the continuum for V=14, and S/N of about 40 at V=16.
The spectra were trimmed, debiased and overscan and flat--field corrected using the IRAF/DOHYDRA package. The wavelength calibration was also performed using this package. All subsequent analysis was performed using the IDL/ICUR package (Walter 1992). The spectra were not flux calibrated. To correct for wavelength dependent detector effects, spectra of three bright stars were combined, heavily smoothed and fitted with a third order polynomial. The fits were then normalized with respect to the maxima. Data analysis were mostly performed by undergraduate student James Petreshock under my supervision.
Equivalent widths of Ca I absorption lines at 6122Å, 6162Å, and
6494Å and Fe I
at 6103Å, 6200Å,
and 6574Å were measured for several spectral standards. All
spectral observations were made with the KPNO 2.1 meter telescope
using the GOLDCAM. The spectral standards were observed several time
per nights so that spectra at different airmasses could be sampled.
The spectral types of the target stars were determined by comparing
source line ratios to those of standard stars.
Line ratios were used
as an indicator of spectral type. The ratios were obtained from the
three Ca I absorption lines: 6122/6494, 6162/6494, and 6122/6162.
As shown in Figures 3 and 4,
the Ca I lines show a strong dependence upon
spectral type over the range of spectral type we were expecting to
observe. The majority of the spectra showed that the
Fe I line at 6574 was filled in by H
emission and therefore would
not be an accurate indicator. Also, the Fe I lines were generally weaker
than the Ca I lines. In Figures 3 and
4, the solid line represents a third order
polynomial fit to the observed data. The standard deviation to this fit is
about 1.5 spectral sub--types. I take this to be the error
in the final spectral type determination of the target stars.
The standard used are given in Table 10
It was not possible to classify sources with an average count rate below 100 photons/pixel. The results of the spectral classification are shown in Tables 12 through 11. The uncertainty in the spectral type is about one and a half sub--types. For stars with S/N of greater than 100, the deviation was usually less than one spectral subtype. However, in the cases of high noise observations, the deviation was up to three subtypes.
All but one of the bright X--ray sources near 
Orionis shows a
Li absorption feature and six (out of 49 = 12.3%) show strong (>10Å)
H
emission lines. Northwest of the belt, again all but one
source shows Li in absorption,
but only one star (out of 13 = 7.7%) shows an H
emission line.
The data verify that the X--ray
emission is a good indicator of youth.
