3 Archival UV and optical data

In order to complement our JHK photometry, and to be able to perform our broad-energy analysis, we have compiled published optical data and UV data from the same sources described in Paper I (see therein for references). Resulting energy distributions are plotted in Fig. 1.

3.1 Optical data

For BD-3 5357 (FF Aqr), Feige 110, HD4539, PG0856+121 and the CSPN BD-13 842 - not analysed, the photometric data collected show offsets among the various systems difficult to interpret in terms of observational errors alone. For some of the stars offsets between sets of photometric data suggest that the cause of the differences is not variability in the sources but systematic differences in the observation and reduction techniques employed by the various authors.

3.2 IUE data

After their extraction from the NASA/RDAF data base at Greenbelt (Maryland) and from the Unified Low Dispersion Archive (ULDA) (Wamsteker et al.$\,$1989) at LAEFF (Madrid), we list the IUE spectra employed in this work in Table 6. The spectra were chosen with the same criteria given in Paper I and were also processed as there described. Reddening corrections - using Seaton's (1979) law, were also applied when required, given the E(B-V) estimates shown in Table 6. We will not discuss line-profile analysis of the H+HeII line at 1215 Å available in the low-dispersion IUE SW spectra collected.

  

Table 6: Table of large-aperture IUE spectra used in this work. "SWP'', "LWP'' and "LWR'' respectively indicate spectra taken with the short and long wavelength primary, and redundant cameras. Results of reddening analysis of suitable IUE spectra are also included. In spectra where reddening is not readily apparent an upper limit of E(B-V) $\leq$ 0.025 is assigned. The uncertainties given are estimated visually and correspond roughly to 2 or 3 sigma intervals. For PHL678 only an approximate value can be given for the large reddening - for smaller values the "2200 Å bump'' is still very clear and for larger values over-correction takes place, yet the correction at E(B-V)=0.2 is not perfect

to