The UV spectra have been taken with the International Ultraviolet Explorer (IUE)
satellite (Boggess et al. 1978a, b, 1989; Fälker et al.\
1989). We have used all spectra taken between 1978 and 1991 from the IUE
Archive, many of these have been taken under the ESA IUE Observatory Programs. In
these 13 years (4903 days) 158 spectra were obtained through the large aperture
(LAP: ) in low resolution mode (
), 90 with the
SWP (Short Wavelength Prime) camera between
, and 68 with the
LWR/P (Long Wavelength Redundant/Prime) cameras between
.
All were reduced with the most recent version of the IUE Image Processing System
(IUESIPS: Clavel 1987). To adjust for shifts in the wavelength
scale due to centering uncertainties in the LAP, all spectra were aligned by
centering the peak of
at
and for
at
. The LWR spectra were
corrected for the sensitivity loss of this camera with time (Clavel et al.\
1988). On the other hand, because the sensitivity degradation of the SWP
and LWP Cameras is small (some 13% over the 13 years) and shows a smooth
variation with wavelength above 1200 Å (Sonneborn & Garhart
1987; Bohlin & Grillmair 1988), its effect on the line
shape and continuum variation is very small in comparison to the overall
variability of F-9 and no correction was applied for it. The IUE
Fine-Error-Sensor (FES) allows optical photometry in a 12'' equivalent square
aperture with an effective wavelength for AGN-type energy distributions of
. The 58 FES measurements have been corrected for the
sensitivity loss with time. Due to FES reference point (the position on the
detector where the photometric measurements are made) changes, two different
corrections had to be applied: one between 1980 and 1989 (Pérez et al.\
1991) and another after 1990 (Pérez 1991).
Rather than listing here again the details for the individual spectra, we
refer the reader for the
observational details to Clavel et al. (1989) and Courvoisier
& Paltani (1992). In most cases the IUE spectra were exposed optimally
for \
and
but those overexposed in the lines (saturation in more than 5
pixels) have not been used for the line decomposition study (SWP 1804, 2215,
4539, 13351, 16559, 16890, 17249, 17446, 17521, 18094, 18891, 21261, 21286,
25825, 26774, 27007, 27008, 27400, 28213,
28307, 28632, 28633, 29545, 29991, 30160, 31176, 34053, 37548, 40179; LWP
2037, 5872,
6201, 6491, 8081, 8562, 9426). Table 1 gives the UV continuum fluxes and FES
counts for the IUE data. Whenever there are several spectra and several FES
measurements the values in Table 1 give the mean value and the rms error
for the fluxes and the FES counts. In the case
of a single FES measurement, we have adopted an error of 6%, corresponding to
the average of the rms deviations of the individual measurements at the epochs when
more than one measurement was available.