The spectra were obtained with the 2.5 m Du Pont telescope at Las
Campanas Observatory (Chile) on May 20 and 21, 1991 (UT date). This
observing time was allocated for studying faint cataclysmic variable
stars, but bad weather forced us to carry out the backup program
consisting of observations of bright Be stars. The stars, selected from
the Supplement to the Bright Star Catalogue (Hoffleit et al. 1983), are
listed in Table 1 (click here). We used the CRAF CCD array of
1024 
1024 12 
pixels, characterized by a gain of 1.24
e
/ADU and a readout noise of 7.7 e
. A grating of 600 l/mm
tilted at an angle of 29
65 yielded a projected spectral range of
6470-8780 Å. A slit width of 0
7 was selected, giving a
spectral resolution of 5.4 Å. The CCD data were reduced with IRAF
standard routines involving bias subtraction, flat field correction
and subtraction of one-dimensional spectra after removing the night
sky. He-Ne-Ar exposures were taken after each object spectrum giving
wavelength calibrations with a typical rms scatter of 0.02 Å.
Non-photometric weather conditions prevented us of obtaining flux
calibrated spectra. Normalized spectra were obtained by fitting the
continuum with a high order polynomial, the resulting data are given in
Table 2 (click here). Equivalent widths (W) were measured between the
points where the line flux was equal to the continuum, using the ``e"
routine in the IRAF ``splot" package. Internal errors were of the
order of 2%. Due to the frequent passage of cirrus, it was difficult
to optimally tune the exposure time for HD124834, HD179419 and
HD203699, a situation that led to some top-saturated H
profiles. For these stars a lower limit for 
was then
obtained. The results are given in Table 3 (click here). In the case of
emission profiles superimposed on absorption wings, equivalent widths
were also measured upon the points defining the minima of the emission
lines. These measurements are indicated in Table 3 (click here) with a
preceding slash. In addition, radial velocities were calculated by
fitting the profiles with a gaussian function. Formal errors, derived
from the rms scatter of the wavelength calibrations, are of the order
of 0.3 km s
while internal errors due to the method of
measurement are of the order of 1 km s
. The radial-velocity
results are listed in Table 4 (click here) and they refer to the Local
Standard of Rest (
km s
, 
and 
).
The stellar radial velocities of Be stars
are commonly derived from the optical helium lines which are not so
contaminated by emission. Due to the lack of such lines in our
spectral region, it was not possible to correct our measurements for the
stellar proper motion. In addition, because the OI 
7773
triplet was not resolved in our spectra, we did not measure radial
velocities for this line. Spectra of HD90966, HD91188 and AG Car
are shown in Figs. 1 (click here), 2 (click here) and
3 (click here) for illustration.
Figure 4 (click here) gives the position of the 7 Be stars in the Geneva (reddening-free) X,Y space, which is the Geneva observational H-R diagram for hot stars (Cramer & Maeder 1979)--the solid lines indicate the locus of the early-type main-sequence stars, giants and (Iab) supergiants, and are taken from Cramer (1994).
Figure 1: Normalized spectrum of HD90966. Note the strong H
emission and the weaker OI emission at 
7775
Figure 2: Normalized spectrum of HD91188. A weak H
emission
and absorptions at HeI, OI and Paschen lines characterize this
spectrum
Figure 3: Normalized spectrum of AG Car. A strong H
emission
peak with broad emission wings dominate the spectrum. Prominent He I
emission lines are also visible
Figure 4: Position of the 7 Be stars in the Geneva X,Y photometric
parameter space, which is the Geneva observational H-R
diagram for hot stars (Cramer & Maeder 1979)--the solid lines
indicate the locus of the early-type main-sequence stars, giants and
(Iab) supergiants, and are taken from Cramer (1994)
Figure 5: H
emission line profiles for AG Car (right) and for
the 7 Be stars on a same intensity scale (tick marks 0.5 apart)
Figure 5: Phase diagram of Strömgren b data from Balona et al.
(1992). The arrow indicates the phase at which our spectrogram was
taken. Note the double wave
Figure 7: Composite V light curve of AG Car based on data from
miscellaneous sources. The vertical arrow denotes the moment of our
observation, the thin line is a hand-drawn curve to illustrate the
approximate magnitude level of the star at that epoch